Publications
Search
1.
Aris, Timothy; Ustun, Volkan; Kumar, Rajay
Learning to Take Cover with Navigation-Based Waypoints via Reinforcement Learning Journal Article
In: FLAIRS, vol. 36, 2023, ISSN: 2334-0762.
@article{aris_learning_2023,
title = {Learning to Take Cover with Navigation-Based Waypoints via Reinforcement Learning},
author = {Timothy Aris and Volkan Ustun and Rajay Kumar},
url = {https://journals.flvc.org/FLAIRS/article/view/133348},
doi = {10.32473/flairs.36.133348},
issn = {2334-0762},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-04},
journal = {FLAIRS},
volume = {36},
abstract = {This paper presents a reinforcement learning model designed to learn how to take cover on geo-specific terrains, an essential behavior component for military training simulations. Training of the models is performed on the Rapid Integration and Development Environment (RIDE) leveraging the Unity ML-Agents framework. This work expands on previous work on raycast-based agents by increasing the number of enemies from one to three. We demonstrate an automated way of generating training and testing data within geo-specific terrains. We show that replacing the action space with a more abstracted, navmesh-based waypoint movement system can increase the generality and success rate of the models while providing similar results to our previous paper's results regarding retraining across terrains. We also comprehensively evaluate the differences between these and the previous models. Finally, we show that incorporating pixels into the model's input can increase performance at the cost of longer training times.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper presents a reinforcement learning model designed to learn how to take cover on geo-specific terrains, an essential behavior component for military training simulations. Training of the models is performed on the Rapid Integration and Development Environment (RIDE) leveraging the Unity ML-Agents framework. This work expands on previous work on raycast-based agents by increasing the number of enemies from one to three. We demonstrate an automated way of generating training and testing data within geo-specific terrains. We show that replacing the action space with a more abstracted, navmesh-based waypoint movement system can increase the generality and success rate of the models while providing similar results to our previous paper's results regarding retraining across terrains. We also comprehensively evaluate the differences between these and the previous models. Finally, we show that incorporating pixels into the model's input can increase performance at the cost of longer training times.
2.
Rosenbloom, Paul S.
Rethinking the Physical Symbol Systems Hypothesis Book Section
In: Hammer, Patrick; Alirezaie, Marjan; Strannegård, Claes (Ed.): Artificial General Intelligence, vol. 13921, pp. 207–216, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-33468-9 978-3-031-33469-6, (Series Title: Lecture Notes in Computer Science).
@incollection{hammer_rethinking_2023,
title = {Rethinking the Physical Symbol Systems Hypothesis},
author = {Paul S. Rosenbloom},
editor = {Patrick Hammer and Marjan Alirezaie and Claes Strannegård},
url = {https://link.springer.com/10.1007/978-3-031-33469-6_21},
doi = {10.1007/978-3-031-33469-6_21},
isbn = {978-3-031-33468-9 978-3-031-33469-6},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-24},
booktitle = {Artificial General Intelligence},
volume = {13921},
pages = {207–216},
publisher = {Springer Nature Switzerland},
address = {Cham},
note = {Series Title: Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
3.
Chadalapaka, Viswanath; Ustun, Volkan; Liu, Lixing
Leveraging Graph Networks to Model Environments in Reinforcement Learning Journal Article
In: FLAIRS, vol. 36, 2023, ISSN: 2334-0762.
@article{chadalapaka_leveraging_2023,
title = {Leveraging Graph Networks to Model Environments in Reinforcement Learning},
author = {Viswanath Chadalapaka and Volkan Ustun and Lixing Liu},
url = {https://journals.flvc.org/FLAIRS/article/view/133118},
doi = {10.32473/flairs.36.133118},
issn = {2334-0762},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-04},
journal = {FLAIRS},
volume = {36},
abstract = {This paper proposes leveraging graph neural networks (GNNs) to model an agent’s environment to construct superior policy networks in reinforcement learning (RL). To this end, we explore the effects of different combinations of GNNs and graph network pooling functions on policy performance. We also run experiments at different levels of problem complexity, which affect how easily we expect an agent to learn an optimal policy and therefore show whether or not graph networks are effective at various problem complexity levels. The efficacy of our approach is shown via experimentation in a partially-observable, non-stationary environment that parallels the highly-practical scenario of a military training exercise with human trainees, where the learning goal is to become the best sparring partner possible for human trainees. Our results present that our models can generate better-performing sparring partners by employing GNNs, as demonstrated by these experiments in the proof-of-concept environment. We also explore our model’s applicability in Multi-Agent RL scenarios. Our code is available online at https://github.com/Derposoft/GNNsAsEnvs.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper proposes leveraging graph neural networks (GNNs) to model an agent’s environment to construct superior policy networks in reinforcement learning (RL). To this end, we explore the effects of different combinations of GNNs and graph network pooling functions on policy performance. We also run experiments at different levels of problem complexity, which affect how easily we expect an agent to learn an optimal policy and therefore show whether or not graph networks are effective at various problem complexity levels. The efficacy of our approach is shown via experimentation in a partially-observable, non-stationary environment that parallels the highly-practical scenario of a military training exercise with human trainees, where the learning goal is to become the best sparring partner possible for human trainees. Our results present that our models can generate better-performing sparring partners by employing GNNs, as demonstrated by these experiments in the proof-of-concept environment. We also explore our model’s applicability in Multi-Agent RL scenarios. Our code is available online at https://github.com/Derposoft/GNNsAsEnvs.
4.
Zhou, Jincheng; Ustun, Volkan
PySigma: Towards Enhanced Grand Unification for the Sigma Cognitive Architecture Book Section
In: Goertzel, Ben; Iklé, Matthew; Potapov, Alexey (Ed.): Artificial General Intelligence, vol. 13154, pp. 355–366, Springer International Publishing, Cham, 2022, ISBN: 978-3-030-93757-7 978-3-030-93758-4.
@incollection{zhou_pysigma_2022,
title = {PySigma: Towards Enhanced Grand Unification for the Sigma Cognitive Architecture},
author = {Jincheng Zhou and Volkan Ustun},
editor = {Ben Goertzel and Matthew Iklé and Alexey Potapov},
url = {https://link.springer.com/10.1007/978-3-030-93758-4_36},
doi = {10.1007/978-3-030-93758-4_36},
isbn = {978-3-030-93757-7 978-3-030-93758-4},
year = {2022},
date = {2022-01-01},
urldate = {2022-09-21},
booktitle = {Artificial General Intelligence},
volume = {13154},
pages = {355–366},
publisher = {Springer International Publishing},
address = {Cham},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
5.
Joshi, Himanshu; Rosenbloom, Paul S.; Ustun, Volkan
Continuous phone recognition in the Sigma cognitive architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 18, pp. 23–32, 2016, ISSN: 2212683X.
@article{joshi_continuous_2016,
title = {Continuous phone recognition in the Sigma cognitive architecture},
author = {Himanshu Joshi and Paul S. Rosenbloom and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X16300652},
doi = {10.1016/j.bica.2016.09.001},
issn = {2212683X},
year = {2016},
date = {2016-10-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {18},
pages = {23–32},
abstract = {Spoken language processing is an important capability of human intelligence that has hitherto been unexplored by cognitive architectures. This reflects on both the symbolic and sub-symbolic nature of the speech problem, and the capabilities provided by cognitive architectures to model the latter and its rich interplay with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. In this article, previous work on speaker dependent isolated word recognition has been extended to demonstrate Sigma’s feasibility to process a stream of fluent audio and recognize phones, in an online and incremental manner with speaker independence. Phone recognition is an important step in integrating spoken language processing into Sigma. This work also extends the acoustic front-end used in the previous work in service of speaker independence. All of the knowledge used in phone recognition was added supraarchitecturally – i.e. on top of the architecture – without requiring the addition of new mechanisms to the architecture.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Spoken language processing is an important capability of human intelligence that has hitherto been unexplored by cognitive architectures. This reflects on both the symbolic and sub-symbolic nature of the speech problem, and the capabilities provided by cognitive architectures to model the latter and its rich interplay with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. In this article, previous work on speaker dependent isolated word recognition has been extended to demonstrate Sigma’s feasibility to process a stream of fluent audio and recognize phones, in an online and incremental manner with speaker independence. Phone recognition is an important step in integrating spoken language processing into Sigma. This work also extends the acoustic front-end used in the previous work in service of speaker independence. All of the knowledge used in phone recognition was added supraarchitecturally – i.e. on top of the architecture – without requiring the addition of new mechanisms to the architecture.
6.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
The Sigma Cognitive Architecture and System: Towards Functionally Elegant Grand Unification Journal Article
In: Journal of Artificial General Intelligence, 2016, ISSN: 1946-0163.
@article{rosenbloom_sigma_2016,
title = {The Sigma Cognitive Architecture and System: Towards Functionally Elegant Grand Unification},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://www.degruyter.com/view/j/jagi.ahead-of-print/jagi-2016-0001/jagi-2016-0001.xml},
doi = {10.1515/jagi-2016-0001},
issn = {1946-0163},
year = {2016},
date = {2016-07-01},
journal = {Journal of Artificial General Intelligence},
abstract = {Sigma (Σ) is a cognitive architecture and system whose development is driven by a combination of four desiderata: grand unification, generic cognition, functional elegance, and sufficient efficiency. Work towards these desiderata is guided by the graphical architecture hypothesis, that key to progress on them is combining what has been learned from over three decades’ worth of separate work on cognitive architectures and graphical models. In this article, these four desiderata are motivated and explained, and then combined with the graphical architecture hypothesis to yield a rationale for the development of Sigma. The current state of the cognitive architecture is then introduced in detail, along with the graphical architecture that sits below it and implements it. Progress in extending Sigma beyond these architectures and towards a full cognitive system is then detailed in terms of both a systematic set of higher level cognitive idioms that have been developed and several virtual humans that are built from combinations of these idioms. Sigma as a whole is then analyzed in terms of how well the progress to date satisfies the desiderata. This article thus provides the first full motivation, presentation and analysis of Sigma, along with a diversity of more specific results that have been generated during its development.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Sigma (Σ) is a cognitive architecture and system whose development is driven by a combination of four desiderata: grand unification, generic cognition, functional elegance, and sufficient efficiency. Work towards these desiderata is guided by the graphical architecture hypothesis, that key to progress on them is combining what has been learned from over three decades’ worth of separate work on cognitive architectures and graphical models. In this article, these four desiderata are motivated and explained, and then combined with the graphical architecture hypothesis to yield a rationale for the development of Sigma. The current state of the cognitive architecture is then introduced in detail, along with the graphical architecture that sits below it and implements it. Progress in extending Sigma beyond these architectures and towards a full cognitive system is then detailed in terms of both a systematic set of higher level cognitive idioms that have been developed and several virtual humans that are built from combinations of these idioms. Sigma as a whole is then analyzed in terms of how well the progress to date satisfies the desiderata. This article thus provides the first full motivation, presentation and analysis of Sigma, along with a diversity of more specific results that have been generated during its development.
7.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
Rethinking Sigma’s Graphical Architecture: An Extension to Neural Networks Proceedings Article
In: International Conference on Artificial General Intelligence, pp. 84–94, Springer, New York, NY, 2016, ISBN: 978-3-319-41649-6.
@inproceedings{rosenbloom_rethinking_2016,
title = {Rethinking Sigma’s Graphical Architecture: An Extension to Neural Networks},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://link.springer.com/chapter/10.1007/978-3-319-41649-6_9},
doi = {10.1007/978-3-319-41649-6_9},
isbn = {978-3-319-41649-6},
year = {2016},
date = {2016-07-01},
booktitle = {International Conference on Artificial General Intelligence},
volume = {9782},
pages = {84–94},
publisher = {Springer},
address = {New York, NY},
abstract = {The status of Sigma’s grounding in graphical models is challenged by the ways in which their semantics has been violated while incorporating rule-based reasoning into them. This has led to a rethinking of what goes on in its graphical architecture, with results that include a straightforward extension to feedforward neural networks (although not yet with learning).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The status of Sigma’s grounding in graphical models is challenged by the ways in which their semantics has been violated while incorporating rule-based reasoning into them. This has led to a rethinking of what goes on in its graphical architecture, with results that include a straightforward extension to feedforward neural networks (although not yet with learning).
8.
Ustun, Volkan; Rosenbloom, Paul
Towards Truly Autonomous Synthetic Characters with the Sigma Cognitive Architecture Book Section
In: Integrating Cognitive Architectures into Virtual Character Design, pp. 213 – 237, IGI Global, Hershey, PA, 2016, ISBN: 978-1-5225-0454-2.
@incollection{ustun_towards_2016,
title = {Towards Truly Autonomous Synthetic Characters with the Sigma Cognitive Architecture},
author = {Volkan Ustun and Paul Rosenbloom},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-0454-2},
isbn = {978-1-5225-0454-2},
year = {2016},
date = {2016-06-01},
booktitle = {Integrating Cognitive Architectures into Virtual Character Design},
pages = {213 – 237},
publisher = {IGI Global},
address = {Hershey, PA},
abstract = {Realism is required not only for how synthetic characters look but also for how they behave. Many applications, such as simulations, virtual worlds, and video games, require computational models of intelligence that generate realistic and credible behavior for the participating synthetic characters. Sigma (Σ) is being built as a computational model of general intelligence with a long-term goal of understanding and replicating the architecture of the mind; i.e., the fixed structure underlying intelligent behavior. Sigma leverages probabilistic graphical models towards a uniform grand unification of not only traditional cognitive capabilities but also key non-cognitive aspects, creating unique opportunities for the construction of new kinds of non-modular behavioral models. These ambitions strive for the complete control of synthetic characters that behave as humanly as possible. In this paper, Sigma is introduced along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Realism is required not only for how synthetic characters look but also for how they behave. Many applications, such as simulations, virtual worlds, and video games, require computational models of intelligence that generate realistic and credible behavior for the participating synthetic characters. Sigma (Σ) is being built as a computational model of general intelligence with a long-term goal of understanding and replicating the architecture of the mind; i.e., the fixed structure underlying intelligent behavior. Sigma leverages probabilistic graphical models towards a uniform grand unification of not only traditional cognitive capabilities but also key non-cognitive aspects, creating unique opportunities for the construction of new kinds of non-modular behavioral models. These ambitions strive for the complete control of synthetic characters that behave as humanly as possible. In this paper, Sigma is introduced along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.
9.
Ustun, Volkan; Rosenbloom, Paul S.; Kim, Julia; Li, Lingshan
BUILDING HIGH FIDELITY HUMAN BEHAVIOR MODELS IN THE SIGMA COGNITIVE ARCHITECTURE Proceedings Article
In: Proceedings of the 2015 Winter Simulation Conference, pp. 3124–3125, IEEE, Huntington Beach, CA, 2015, ISBN: 978-1-4673-9741-4.
@inproceedings{ustun_building_2015,
title = {BUILDING HIGH FIDELITY HUMAN BEHAVIOR MODELS IN THE SIGMA COGNITIVE ARCHITECTURE},
author = {Volkan Ustun and Paul S. Rosenbloom and Julia Kim and Lingshan Li},
url = {http://dl.acm.org/citation.cfm?id=2888619.2888999},
isbn = {978-1-4673-9741-4},
year = {2015},
date = {2015-12-01},
booktitle = {Proceedings of the 2015 Winter Simulation Conference},
pages = {3124–3125},
publisher = {IEEE},
address = {Huntington Beach, CA},
abstract = {Many agent simulations involve computational models of intelligent human behavior. In a variety of cases, these behavior models should be high-fidelity to provide the required realism and credibility. Cognitive architectures may assist the generation of such high-fidelity models as they specify the fixed structure underlying an intelligent cognitive system that does not change over time and across domains. Existing symbolic architectures, such as Soar and ACT-R, have been used in this way, but here the focus is on a new architecture, Sigma (!), that leverages probabilistic graphical models towards a uniform grand unification of not only the traditional cognitive capabilities but also key non-cognitive aspects, and which thus yields unique opportunities for construction of new kinds of non-modular high-fidelity behavior models. Here, we briefly introduce Sigma along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Many agent simulations involve computational models of intelligent human behavior. In a variety of cases, these behavior models should be high-fidelity to provide the required realism and credibility. Cognitive architectures may assist the generation of such high-fidelity models as they specify the fixed structure underlying an intelligent cognitive system that does not change over time and across domains. Existing symbolic architectures, such as Soar and ACT-R, have been used in this way, but here the focus is on a new architecture, Sigma (!), that leverages probabilistic graphical models towards a uniform grand unification of not only the traditional cognitive capabilities but also key non-cognitive aspects, and which thus yields unique opportunities for construction of new kinds of non-modular high-fidelity behavior models. Here, we briefly introduce Sigma along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.
10.
Ustun, Volkan; Rosenbloom, Paul S.
Towards Adaptive, Interactive Virtual Humans in Sigma Proceedings Article
In: Intelligent Virtual Agents, pp. 98 –108, Springer, Delft, Netherlands, 2015, ISBN: 978-3-319-21995-0.
@inproceedings{ustun_towards_2015,
title = {Towards Adaptive, Interactive Virtual Humans in Sigma},
author = {Volkan Ustun and Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20Adaptive,%20Interactive%20Virtual%20Humans%20in%20Sigma.pdf},
doi = {10.1007/978-3-319-21996-7_10},
isbn = {978-3-319-21995-0},
year = {2015},
date = {2015-08-01},
booktitle = {Intelligent Virtual Agents},
volume = {9238},
pages = {98 –108},
publisher = {Springer},
address = {Delft, Netherlands},
abstract = {Sigma is a nascent cognitive architecture/system that combines concepts from graphical models with traditional symbolic architectures. Here an initial Sigma-based virtual human (VH) is introduced that combines probabilistic reasoning, rule-based decision-making, Theory of Mind, Simultaneous Localization and Mapping and reinforcement learning in a unified manner. This non-modular unification of diverse cognitive, robotic and VH capabilities provides an important first step towards fully adaptive and interactive VHs in Sigma.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Sigma is a nascent cognitive architecture/system that combines concepts from graphical models with traditional symbolic architectures. Here an initial Sigma-based virtual human (VH) is introduced that combines probabilistic reasoning, rule-based decision-making, Theory of Mind, Simultaneous Localization and Mapping and reinforcement learning in a unified manner. This non-modular unification of diverse cognitive, robotic and VH capabilities provides an important first step towards fully adaptive and interactive VHs in Sigma.
11.
Rosenbloom, Paul S.; Gratch, Jonathan; Ustun, Volkan
Towards Emotion in Sigma: From Appraisal to Attention Proceedings Article
In: Proceedings of AGI 2015, pp. 142 – 151, Springer International Publishing, Berlin, Germany, 2015.
@inproceedings{rosenbloom_towards_2015,
title = {Towards Emotion in Sigma: From Appraisal to Attention},
author = {Paul S. Rosenbloom and Jonathan Gratch and Volkan Ustun},
url = {http://ict.usc.edu/pubs/Towards%20Emotion%20in%20Sigma%20-%20From%20Appraisal%20to%20Attention.pdf},
year = {2015},
date = {2015-07-01},
booktitle = {Proceedings of AGI 2015},
volume = {9205},
pages = {142 – 151},
publisher = {Springer International Publishing},
address = {Berlin, Germany},
abstract = {A first step is taken towards incorporating emotional processing into Sigma, a cognitive architecture that is grounded in graphical models, with the addition of appraisal variables for expectedness and desirability plus their initial implications for attention at two levels of the control hierarchy. The results leverage many of Sigma's existing capabilities but with a few key additions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A first step is taken towards incorporating emotional processing into Sigma, a cognitive architecture that is grounded in graphical models, with the addition of appraisal variables for expectedness and desirability plus their initial implications for attention at two levels of the control hierarchy. The results leverage many of Sigma's existing capabilities but with a few key additions.
12.
Kommers, Cody; Ustun, Volkan; Demski, Abram; Rosenbloom, Paul
Hierarchical Reasoning with Distributed Vector Representations Proceedings Article
In: Proceedings of 37th Annual Conference of the Cognitive Science Society, Cognitive Science Society, Pasadena, CA, 2015.
@inproceedings{kommers_hierarchical_2015,
title = {Hierarchical Reasoning with Distributed Vector Representations},
author = {Cody Kommers and Volkan Ustun and Abram Demski and Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Hierarchical%20Reasoning%20with%20Distributed%20Vector%20Representations.pdf},
year = {2015},
date = {2015-07-01},
booktitle = {Proceedings of 37th Annual Conference of the Cognitive Science Society},
publisher = {Cognitive Science Society},
address = {Pasadena, CA},
abstract = {We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.
13.
Rosenbloom, Paul
Supraarchitectural Capability Integration: From Soar to Sigma Proceedings Article
In: Proceedings of the 13th International Conference on Cognitive Modeling, 2015, Groningen, The Netherlands, 2015.
@inproceedings{rosenbloom_supraarchitectural_2015,
title = {Supraarchitectural Capability Integration: From Soar to Sigma},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Supraarchitectural%20Capability%20Integration%20-%20From%20Soar%20to%20Sigma.pdf},
year = {2015},
date = {2015-04-01},
booktitle = {Proceedings of the 13th International Conference on Cognitive Modeling, 2015},
address = {Groningen, The Netherlands},
abstract = {Integration across capabilities, both architectural and supraarchitectural, is critical for cognitive architectures. Here we revisit a classic failure of supraarchitectural capability integration in Soar, involving data chunking, to understand better both its source and how it and related integration issues can be overcome via three general extensions in Sigma.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Integration across capabilities, both architectural and supraarchitectural, is critical for cognitive architectures. Here we revisit a classic failure of supraarchitectural capability integration in Soar, involving data chunking, to understand better both its source and how it and related integration issues can be overcome via three general extensions in Sigma.
14.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
Efficient message computation in Sigma’s graphical architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 11, pp. 1–9, 2015, ISSN: 2212683X.
@article{rosenbloom_efficient_2015,
title = {Efficient message computation in Sigma’s graphical architecture},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X14000723},
doi = {10.1016/j.bica.2014.11.009},
issn = {2212683X},
year = {2015},
date = {2015-01-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {11},
pages = {1–9},
abstract = {Human cognition runs at ∼50 ms per cognitive cycle, implying that any biologically inspired cognitive architecture that strives for real-time performance needs to be able to run at this speed. Sigma is a cognitive architecture built upon graphical models – a broadly applicable state-of-the-art formalism for implementing cognitive capabilities – that are solved via message passing (with complex messages based on n-dimensional piecewise-linear functions). Earlier work explored optimizations to Sigma that reduced by an order of magnitude the number of messages sent per cycle. Here, optimizations are introduced that reduce by an order of magnitude the average time required per message sent.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Human cognition runs at ∼50 ms per cognitive cycle, implying that any biologically inspired cognitive architecture that strives for real-time performance needs to be able to run at this speed. Sigma is a cognitive architecture built upon graphical models – a broadly applicable state-of-the-art formalism for implementing cognitive capabilities – that are solved via message passing (with complex messages based on n-dimensional piecewise-linear functions). Earlier work explored optimizations to Sigma that reduced by an order of magnitude the number of messages sent per cycle. Here, optimizations are introduced that reduce by an order of magnitude the average time required per message sent.
15.
Joshi, Himanshu; Rosenbloom, Paul S.; Ustun, Volkan
Isolated word recognition in the Sigma cognitive architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 10, pp. 1–9, 2014, ISSN: 2212683X.
@article{joshi_isolated_2014,
title = {Isolated word recognition in the Sigma cognitive architecture},
author = {Himanshu Joshi and Paul S. Rosenbloom and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X14000644},
doi = {10.1016/j.bica.2014.11.001},
issn = {2212683X},
year = {2014},
date = {2014-10-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {10},
pages = {1–9},
abstract = {Symbolic architectures are effective at complex cognitive reasoning, but typically are incapable of important forms of sub-cognitive processing – such as perception – without distinct modules connected to them via low-bandwidth interfaces. Neural architectures, in contrast, may be quite effective at the latter, but typically struggle with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. Here it is shown that Sigma is not only capable of performing a simple variant of speech recognition via the same knowledge structures and reasoning algorithm used for cognitive processing, but also of leveraging its existing knowledge templates and learning algorithm to acquire automatically most of the structures and parameters needed for this recognition activity.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Symbolic architectures are effective at complex cognitive reasoning, but typically are incapable of important forms of sub-cognitive processing – such as perception – without distinct modules connected to them via low-bandwidth interfaces. Neural architectures, in contrast, may be quite effective at the latter, but typically struggle with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. Here it is shown that Sigma is not only capable of performing a simple variant of speech recognition via the same knowledge structures and reasoning algorithm used for cognitive processing, but also of leveraging its existing knowledge templates and learning algorithm to acquire automatically most of the structures and parameters needed for this recognition activity.
16.
Ustun, Volkan; Rosenbloom, Paul S.; Sagae, Kenji; Demski, Abram
Distributed Vector Representations of Words in the Sigma Cognitive Architecture Proceedings Article
In: Proceedings of the 7th Conference on Artificial General Intelligence 2014, Québec City, Canada, 2014.
@inproceedings{ustun_distributed_2014,
title = {Distributed Vector Representations of Words in the Sigma Cognitive Architecture},
author = {Volkan Ustun and Paul S. Rosenbloom and Kenji Sagae and Abram Demski},
url = {http://ict.usc.edu/pubs/Distributed%20Vector%20Representations%20of%20Words%20in%20the%20Sigma%20Cognitive%20Architecture.pdf},
year = {2014},
date = {2014-08-01},
booktitle = {Proceedings of the 7th Conference on Artificial General Intelligence 2014},
address = {Québec City, Canada},
abstract = {Recently reported results with distributed-vector word representations in natural language processing make them appealing for incorporation into a general cognitive architecture like Sigma. This paper describes a new algorithm for learning such word representations from large, shallow information resources, and how this algorithm can be implemented via small modifications to Sigma. The effectiveness and speed of the algorithm are evaluated via a comparison of an external simulation of it with state-of-the-art algorithms. The results from more limited experiments with Sigma are also promising, but more work is required for it to reach the effectiveness and speed of the simulation.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Recently reported results with distributed-vector word representations in natural language processing make them appealing for incorporation into a general cognitive architecture like Sigma. This paper describes a new algorithm for learning such word representations from large, shallow information resources, and how this algorithm can be implemented via small modifications to Sigma. The effectiveness and speed of the algorithm are evaluated via a comparison of an external simulation of it with state-of-the-art algorithms. The results from more limited experiments with Sigma are also promising, but more work is required for it to reach the effectiveness and speed of the simulation.
17.
Rosenbloom, Paul S.
Ubiquity symposium: The science in computer science: the computing sciences and STEM education Journal Article
In: Ubiquity, no. March 2014, pp. 1–8, 2014, ISSN: 15302180.
@article{rosenbloom_ubiquity_2014,
title = {Ubiquity symposium: The science in computer science: the computing sciences and STEM education},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Ubiquity%20symposium%20-%20The%20science%20in%20computer%20science%20-%20the%20computing%20sciences%20and%20STEM%20education.pdf},
doi = {10.1145/2590528.2590530},
issn = {15302180},
year = {2014},
date = {2014-03-01},
journal = {Ubiquity},
number = {March 2014},
pages = {1–8},
abstract = {In this latest installment of "The Science in Computer Science," Prof. Paul Rosenbloom continues the discussion on whether or not computer science can be considered a "natural science." He argues not only is computing the basis for a true science, it is in fact an entire scientific domain.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this latest installment of "The Science in Computer Science," Prof. Paul Rosenbloom continues the discussion on whether or not computer science can be considered a "natural science." He argues not only is computing the basis for a true science, it is in fact an entire scientific domain.
18.
Pynadath, David V.; Rosenbloom, Paul; Marsella, Stacy C.; Li, Lingshan
Modeling Two-Player Games in the Sigma Graphical Cognitive Architecture Proceedings Article
In: Proceedings of the 6th Conference on Artificial General Intelligence, pp. 98–108, Beijing, China, 2013.
@inproceedings{pynadath_modeling_2013,
title = {Modeling Two-Player Games in the Sigma Graphical Cognitive Architecture},
author = {David V. Pynadath and Paul Rosenbloom and Stacy C. Marsella and Lingshan Li},
url = {http://ict.usc.edu/pubs/Modeling%20Two-Player%20Games%20in%20the%20Sigma%20Graphical%20Cognitive%20Architecture.pdf},
year = {2013},
date = {2013-07-01},
booktitle = {Proceedings of the 6th Conference on Artificial General Intelligence},
pages = {98–108},
address = {Beijing, China},
abstract = {Effective social interaction and, in particular, a Theory of Mind are critical components of human intelligence, allowing us to form beliefs about other people, generate expectations about their behavior, and use those expectations to inform our own decision-making. This article presents an investigation into methods for realizing Theory of Mind within Sigma, a graphical cognitive architecture. By extending the architecture to capture independent decisions and problem-solving for multiple agents, we implemented Sigma models of several canonical examples from game theory. We show that the resulting Sigma agents can capture the same behaviors prescribed by equilibrium solutions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Effective social interaction and, in particular, a Theory of Mind are critical components of human intelligence, allowing us to form beliefs about other people, generate expectations about their behavior, and use those expectations to inform our own decision-making. This article presents an investigation into methods for realizing Theory of Mind within Sigma, a graphical cognitive architecture. By extending the architecture to capture independent decisions and problem-solving for multiple agents, we implemented Sigma models of several canonical examples from game theory. We show that the resulting Sigma agents can capture the same behaviors prescribed by equilibrium solutions.
19.
Rosenbloom, Paul S.; Demski, Abram; Han, Teawon; Ustun, Volkan
Learning via Gradient Descent in Sigma Proceedings Article
In: International Conference on Cognitive Modeling, Ottawa, Canada, 2013.
@inproceedings{rosenbloom_learning_2013,
title = {Learning via Gradient Descent in Sigma},
author = {Paul S. Rosenbloom and Abram Demski and Teawon Han and Volkan Ustun},
url = {http://ict.usc.edu/pubs/Learning%20via%20Gradient%20Descent%20in%20Sigma.pdf},
year = {2013},
date = {2013-07-01},
booktitle = {International Conference on Cognitive Modeling},
address = {Ottawa, Canada},
abstract = {Integrating a gradient-descent learning mechanism at the core of the graphical models upon which the Sigma cognitive architecture/system is built yields learning behaviors that span important forms of both procedural learning (e.g., action and reinforcement learning) and declarative learning (e.g., supervised and unsupervised concept formation), plus several additional forms of learning (e.g., distribution tracking and map learning) relevant to cognitive systems/modeling. The core result presented here is this breadth of cognitive learning behaviors that is producible in this uniform manner.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Integrating a gradient-descent learning mechanism at the core of the graphical models upon which the Sigma cognitive architecture/system is built yields learning behaviors that span important forms of both procedural learning (e.g., action and reinforcement learning) and declarative learning (e.g., supervised and unsupervised concept formation), plus several additional forms of learning (e.g., distribution tracking and map learning) relevant to cognitive systems/modeling. The core result presented here is this breadth of cognitive learning behaviors that is producible in this uniform manner.
20.
Rosenbloom, Paul S.
The Sigma cognitive architecture and system Journal Article
In: AISB Quarterly, pp. 4–13, 2013.
@article{rosenbloom_sigma_2013,
title = {The Sigma cognitive architecture and system},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/The%20Sigma%20cognitive%20architecture%20and%20system.pdf},
year = {2013},
date = {2013-05-01},
journal = {AISB Quarterly},
pages = {4–13},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Filter
2023
Aris, Timothy; Ustun, Volkan; Kumar, Rajay
Learning to Take Cover with Navigation-Based Waypoints via Reinforcement Learning Journal Article
In: FLAIRS, vol. 36, 2023, ISSN: 2334-0762.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, DTIC, UARC, Virtual Humans
@article{aris_learning_2023,
title = {Learning to Take Cover with Navigation-Based Waypoints via Reinforcement Learning},
author = {Timothy Aris and Volkan Ustun and Rajay Kumar},
url = {https://journals.flvc.org/FLAIRS/article/view/133348},
doi = {10.32473/flairs.36.133348},
issn = {2334-0762},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-04},
journal = {FLAIRS},
volume = {36},
abstract = {This paper presents a reinforcement learning model designed to learn how to take cover on geo-specific terrains, an essential behavior component for military training simulations. Training of the models is performed on the Rapid Integration and Development Environment (RIDE) leveraging the Unity ML-Agents framework. This work expands on previous work on raycast-based agents by increasing the number of enemies from one to three. We demonstrate an automated way of generating training and testing data within geo-specific terrains. We show that replacing the action space with a more abstracted, navmesh-based waypoint movement system can increase the generality and success rate of the models while providing similar results to our previous paper's results regarding retraining across terrains. We also comprehensively evaluate the differences between these and the previous models. Finally, we show that incorporating pixels into the model's input can increase performance at the cost of longer training times.},
keywords = {CogArch, Cognitive Architecture, DTIC, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
This paper presents a reinforcement learning model designed to learn how to take cover on geo-specific terrains, an essential behavior component for military training simulations. Training of the models is performed on the Rapid Integration and Development Environment (RIDE) leveraging the Unity ML-Agents framework. This work expands on previous work on raycast-based agents by increasing the number of enemies from one to three. We demonstrate an automated way of generating training and testing data within geo-specific terrains. We show that replacing the action space with a more abstracted, navmesh-based waypoint movement system can increase the generality and success rate of the models while providing similar results to our previous paper's results regarding retraining across terrains. We also comprehensively evaluate the differences between these and the previous models. Finally, we show that incorporating pixels into the model's input can increase performance at the cost of longer training times.
Rosenbloom, Paul S.
Rethinking the Physical Symbol Systems Hypothesis Book Section
In: Hammer, Patrick; Alirezaie, Marjan; Strannegård, Claes (Ed.): Artificial General Intelligence, vol. 13921, pp. 207–216, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-33468-9 978-3-031-33469-6, (Series Title: Lecture Notes in Computer Science).
Links | BibTeX | Tags: CogArch, Cognitive Architecture
@incollection{hammer_rethinking_2023,
title = {Rethinking the Physical Symbol Systems Hypothesis},
author = {Paul S. Rosenbloom},
editor = {Patrick Hammer and Marjan Alirezaie and Claes Strannegård},
url = {https://link.springer.com/10.1007/978-3-031-33469-6_21},
doi = {10.1007/978-3-031-33469-6_21},
isbn = {978-3-031-33468-9 978-3-031-33469-6},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-24},
booktitle = {Artificial General Intelligence},
volume = {13921},
pages = {207–216},
publisher = {Springer Nature Switzerland},
address = {Cham},
note = {Series Title: Lecture Notes in Computer Science},
keywords = {CogArch, Cognitive Architecture},
pubstate = {published},
tppubtype = {incollection}
}
Chadalapaka, Viswanath; Ustun, Volkan; Liu, Lixing
Leveraging Graph Networks to Model Environments in Reinforcement Learning Journal Article
In: FLAIRS, vol. 36, 2023, ISSN: 2334-0762.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, DTIC, UARC
@article{chadalapaka_leveraging_2023,
title = {Leveraging Graph Networks to Model Environments in Reinforcement Learning},
author = {Viswanath Chadalapaka and Volkan Ustun and Lixing Liu},
url = {https://journals.flvc.org/FLAIRS/article/view/133118},
doi = {10.32473/flairs.36.133118},
issn = {2334-0762},
year = {2023},
date = {2023-05-01},
urldate = {2023-08-04},
journal = {FLAIRS},
volume = {36},
abstract = {This paper proposes leveraging graph neural networks (GNNs) to model an agent’s environment to construct superior policy networks in reinforcement learning (RL). To this end, we explore the effects of different combinations of GNNs and graph network pooling functions on policy performance. We also run experiments at different levels of problem complexity, which affect how easily we expect an agent to learn an optimal policy and therefore show whether or not graph networks are effective at various problem complexity levels. The efficacy of our approach is shown via experimentation in a partially-observable, non-stationary environment that parallels the highly-practical scenario of a military training exercise with human trainees, where the learning goal is to become the best sparring partner possible for human trainees. Our results present that our models can generate better-performing sparring partners by employing GNNs, as demonstrated by these experiments in the proof-of-concept environment. We also explore our model’s applicability in Multi-Agent RL scenarios. Our code is available online at https://github.com/Derposoft/GNNsAsEnvs.},
keywords = {CogArch, Cognitive Architecture, DTIC, UARC},
pubstate = {published},
tppubtype = {article}
}
This paper proposes leveraging graph neural networks (GNNs) to model an agent’s environment to construct superior policy networks in reinforcement learning (RL). To this end, we explore the effects of different combinations of GNNs and graph network pooling functions on policy performance. We also run experiments at different levels of problem complexity, which affect how easily we expect an agent to learn an optimal policy and therefore show whether or not graph networks are effective at various problem complexity levels. The efficacy of our approach is shown via experimentation in a partially-observable, non-stationary environment that parallels the highly-practical scenario of a military training exercise with human trainees, where the learning goal is to become the best sparring partner possible for human trainees. Our results present that our models can generate better-performing sparring partners by employing GNNs, as demonstrated by these experiments in the proof-of-concept environment. We also explore our model’s applicability in Multi-Agent RL scenarios. Our code is available online at https://github.com/Derposoft/GNNsAsEnvs.
2022
Zhou, Jincheng; Ustun, Volkan
PySigma: Towards Enhanced Grand Unification for the Sigma Cognitive Architecture Book Section
In: Goertzel, Ben; Iklé, Matthew; Potapov, Alexey (Ed.): Artificial General Intelligence, vol. 13154, pp. 355–366, Springer International Publishing, Cham, 2022, ISBN: 978-3-030-93757-7 978-3-030-93758-4.
Links | BibTeX | Tags: CogArch, Cognitive Architecture, DTIC, UARC
@incollection{zhou_pysigma_2022,
title = {PySigma: Towards Enhanced Grand Unification for the Sigma Cognitive Architecture},
author = {Jincheng Zhou and Volkan Ustun},
editor = {Ben Goertzel and Matthew Iklé and Alexey Potapov},
url = {https://link.springer.com/10.1007/978-3-030-93758-4_36},
doi = {10.1007/978-3-030-93758-4_36},
isbn = {978-3-030-93757-7 978-3-030-93758-4},
year = {2022},
date = {2022-01-01},
urldate = {2022-09-21},
booktitle = {Artificial General Intelligence},
volume = {13154},
pages = {355–366},
publisher = {Springer International Publishing},
address = {Cham},
keywords = {CogArch, Cognitive Architecture, DTIC, UARC},
pubstate = {published},
tppubtype = {incollection}
}
2016
Joshi, Himanshu; Rosenbloom, Paul S.; Ustun, Volkan
Continuous phone recognition in the Sigma cognitive architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 18, pp. 23–32, 2016, ISSN: 2212683X.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@article{joshi_continuous_2016,
title = {Continuous phone recognition in the Sigma cognitive architecture},
author = {Himanshu Joshi and Paul S. Rosenbloom and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X16300652},
doi = {10.1016/j.bica.2016.09.001},
issn = {2212683X},
year = {2016},
date = {2016-10-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {18},
pages = {23–32},
abstract = {Spoken language processing is an important capability of human intelligence that has hitherto been unexplored by cognitive architectures. This reflects on both the symbolic and sub-symbolic nature of the speech problem, and the capabilities provided by cognitive architectures to model the latter and its rich interplay with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. In this article, previous work on speaker dependent isolated word recognition has been extended to demonstrate Sigma’s feasibility to process a stream of fluent audio and recognize phones, in an online and incremental manner with speaker independence. Phone recognition is an important step in integrating spoken language processing into Sigma. This work also extends the acoustic front-end used in the previous work in service of speaker independence. All of the knowledge used in phone recognition was added supraarchitecturally – i.e. on top of the architecture – without requiring the addition of new mechanisms to the architecture.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
Spoken language processing is an important capability of human intelligence that has hitherto been unexplored by cognitive architectures. This reflects on both the symbolic and sub-symbolic nature of the speech problem, and the capabilities provided by cognitive architectures to model the latter and its rich interplay with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. In this article, previous work on speaker dependent isolated word recognition has been extended to demonstrate Sigma’s feasibility to process a stream of fluent audio and recognize phones, in an online and incremental manner with speaker independence. Phone recognition is an important step in integrating spoken language processing into Sigma. This work also extends the acoustic front-end used in the previous work in service of speaker independence. All of the knowledge used in phone recognition was added supraarchitecturally – i.e. on top of the architecture – without requiring the addition of new mechanisms to the architecture.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
The Sigma Cognitive Architecture and System: Towards Functionally Elegant Grand Unification Journal Article
In: Journal of Artificial General Intelligence, 2016, ISSN: 1946-0163.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@article{rosenbloom_sigma_2016,
title = {The Sigma Cognitive Architecture and System: Towards Functionally Elegant Grand Unification},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://www.degruyter.com/view/j/jagi.ahead-of-print/jagi-2016-0001/jagi-2016-0001.xml},
doi = {10.1515/jagi-2016-0001},
issn = {1946-0163},
year = {2016},
date = {2016-07-01},
journal = {Journal of Artificial General Intelligence},
abstract = {Sigma (Σ) is a cognitive architecture and system whose development is driven by a combination of four desiderata: grand unification, generic cognition, functional elegance, and sufficient efficiency. Work towards these desiderata is guided by the graphical architecture hypothesis, that key to progress on them is combining what has been learned from over three decades’ worth of separate work on cognitive architectures and graphical models. In this article, these four desiderata are motivated and explained, and then combined with the graphical architecture hypothesis to yield a rationale for the development of Sigma. The current state of the cognitive architecture is then introduced in detail, along with the graphical architecture that sits below it and implements it. Progress in extending Sigma beyond these architectures and towards a full cognitive system is then detailed in terms of both a systematic set of higher level cognitive idioms that have been developed and several virtual humans that are built from combinations of these idioms. Sigma as a whole is then analyzed in terms of how well the progress to date satisfies the desiderata. This article thus provides the first full motivation, presentation and analysis of Sigma, along with a diversity of more specific results that have been generated during its development.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
Sigma (Σ) is a cognitive architecture and system whose development is driven by a combination of four desiderata: grand unification, generic cognition, functional elegance, and sufficient efficiency. Work towards these desiderata is guided by the graphical architecture hypothesis, that key to progress on them is combining what has been learned from over three decades’ worth of separate work on cognitive architectures and graphical models. In this article, these four desiderata are motivated and explained, and then combined with the graphical architecture hypothesis to yield a rationale for the development of Sigma. The current state of the cognitive architecture is then introduced in detail, along with the graphical architecture that sits below it and implements it. Progress in extending Sigma beyond these architectures and towards a full cognitive system is then detailed in terms of both a systematic set of higher level cognitive idioms that have been developed and several virtual humans that are built from combinations of these idioms. Sigma as a whole is then analyzed in terms of how well the progress to date satisfies the desiderata. This article thus provides the first full motivation, presentation and analysis of Sigma, along with a diversity of more specific results that have been generated during its development.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
Rethinking Sigma’s Graphical Architecture: An Extension to Neural Networks Proceedings Article
In: International Conference on Artificial General Intelligence, pp. 84–94, Springer, New York, NY, 2016, ISBN: 978-3-319-41649-6.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@inproceedings{rosenbloom_rethinking_2016,
title = {Rethinking Sigma’s Graphical Architecture: An Extension to Neural Networks},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://link.springer.com/chapter/10.1007/978-3-319-41649-6_9},
doi = {10.1007/978-3-319-41649-6_9},
isbn = {978-3-319-41649-6},
year = {2016},
date = {2016-07-01},
booktitle = {International Conference on Artificial General Intelligence},
volume = {9782},
pages = {84–94},
publisher = {Springer},
address = {New York, NY},
abstract = {The status of Sigma’s grounding in graphical models is challenged by the ways in which their semantics has been violated while incorporating rule-based reasoning into them. This has led to a rethinking of what goes on in its graphical architecture, with results that include a straightforward extension to feedforward neural networks (although not yet with learning).},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
The status of Sigma’s grounding in graphical models is challenged by the ways in which their semantics has been violated while incorporating rule-based reasoning into them. This has led to a rethinking of what goes on in its graphical architecture, with results that include a straightforward extension to feedforward neural networks (although not yet with learning).
Ustun, Volkan; Rosenbloom, Paul
Towards Truly Autonomous Synthetic Characters with the Sigma Cognitive Architecture Book Section
In: Integrating Cognitive Architectures into Virtual Character Design, pp. 213 – 237, IGI Global, Hershey, PA, 2016, ISBN: 978-1-5225-0454-2.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@incollection{ustun_towards_2016,
title = {Towards Truly Autonomous Synthetic Characters with the Sigma Cognitive Architecture},
author = {Volkan Ustun and Paul Rosenbloom},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-0454-2},
isbn = {978-1-5225-0454-2},
year = {2016},
date = {2016-06-01},
booktitle = {Integrating Cognitive Architectures into Virtual Character Design},
pages = {213 – 237},
publisher = {IGI Global},
address = {Hershey, PA},
abstract = {Realism is required not only for how synthetic characters look but also for how they behave. Many applications, such as simulations, virtual worlds, and video games, require computational models of intelligence that generate realistic and credible behavior for the participating synthetic characters. Sigma (Σ) is being built as a computational model of general intelligence with a long-term goal of understanding and replicating the architecture of the mind; i.e., the fixed structure underlying intelligent behavior. Sigma leverages probabilistic graphical models towards a uniform grand unification of not only traditional cognitive capabilities but also key non-cognitive aspects, creating unique opportunities for the construction of new kinds of non-modular behavioral models. These ambitions strive for the complete control of synthetic characters that behave as humanly as possible. In this paper, Sigma is introduced along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {incollection}
}
Realism is required not only for how synthetic characters look but also for how they behave. Many applications, such as simulations, virtual worlds, and video games, require computational models of intelligence that generate realistic and credible behavior for the participating synthetic characters. Sigma (Σ) is being built as a computational model of general intelligence with a long-term goal of understanding and replicating the architecture of the mind; i.e., the fixed structure underlying intelligent behavior. Sigma leverages probabilistic graphical models towards a uniform grand unification of not only traditional cognitive capabilities but also key non-cognitive aspects, creating unique opportunities for the construction of new kinds of non-modular behavioral models. These ambitions strive for the complete control of synthetic characters that behave as humanly as possible. In this paper, Sigma is introduced along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.
2015
Ustun, Volkan; Rosenbloom, Paul S.; Kim, Julia; Li, Lingshan
BUILDING HIGH FIDELITY HUMAN BEHAVIOR MODELS IN THE SIGMA COGNITIVE ARCHITECTURE Proceedings Article
In: Proceedings of the 2015 Winter Simulation Conference, pp. 3124–3125, IEEE, Huntington Beach, CA, 2015, ISBN: 978-1-4673-9741-4.
Abstract | Links | BibTeX | Tags: CogArch, Virtual Humans
@inproceedings{ustun_building_2015,
title = {BUILDING HIGH FIDELITY HUMAN BEHAVIOR MODELS IN THE SIGMA COGNITIVE ARCHITECTURE},
author = {Volkan Ustun and Paul S. Rosenbloom and Julia Kim and Lingshan Li},
url = {http://dl.acm.org/citation.cfm?id=2888619.2888999},
isbn = {978-1-4673-9741-4},
year = {2015},
date = {2015-12-01},
booktitle = {Proceedings of the 2015 Winter Simulation Conference},
pages = {3124–3125},
publisher = {IEEE},
address = {Huntington Beach, CA},
abstract = {Many agent simulations involve computational models of intelligent human behavior. In a variety of cases, these behavior models should be high-fidelity to provide the required realism and credibility. Cognitive architectures may assist the generation of such high-fidelity models as they specify the fixed structure underlying an intelligent cognitive system that does not change over time and across domains. Existing symbolic architectures, such as Soar and ACT-R, have been used in this way, but here the focus is on a new architecture, Sigma (!), that leverages probabilistic graphical models towards a uniform grand unification of not only the traditional cognitive capabilities but also key non-cognitive aspects, and which thus yields unique opportunities for construction of new kinds of non-modular high-fidelity behavior models. Here, we briefly introduce Sigma along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.},
keywords = {CogArch, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Many agent simulations involve computational models of intelligent human behavior. In a variety of cases, these behavior models should be high-fidelity to provide the required realism and credibility. Cognitive architectures may assist the generation of such high-fidelity models as they specify the fixed structure underlying an intelligent cognitive system that does not change over time and across domains. Existing symbolic architectures, such as Soar and ACT-R, have been used in this way, but here the focus is on a new architecture, Sigma (!), that leverages probabilistic graphical models towards a uniform grand unification of not only the traditional cognitive capabilities but also key non-cognitive aspects, and which thus yields unique opportunities for construction of new kinds of non-modular high-fidelity behavior models. Here, we briefly introduce Sigma along with two disparate proof-of-concept virtual humans – one conversational and the other a pair of ambulatory agents – that demonstrate its diverse capabilities.
Ustun, Volkan; Rosenbloom, Paul S.
Towards Adaptive, Interactive Virtual Humans in Sigma Proceedings Article
In: Intelligent Virtual Agents, pp. 98 –108, Springer, Delft, Netherlands, 2015, ISBN: 978-3-319-21995-0.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@inproceedings{ustun_towards_2015,
title = {Towards Adaptive, Interactive Virtual Humans in Sigma},
author = {Volkan Ustun and Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20Adaptive,%20Interactive%20Virtual%20Humans%20in%20Sigma.pdf},
doi = {10.1007/978-3-319-21996-7_10},
isbn = {978-3-319-21995-0},
year = {2015},
date = {2015-08-01},
booktitle = {Intelligent Virtual Agents},
volume = {9238},
pages = {98 –108},
publisher = {Springer},
address = {Delft, Netherlands},
abstract = {Sigma is a nascent cognitive architecture/system that combines concepts from graphical models with traditional symbolic architectures. Here an initial Sigma-based virtual human (VH) is introduced that combines probabilistic reasoning, rule-based decision-making, Theory of Mind, Simultaneous Localization and Mapping and reinforcement learning in a unified manner. This non-modular unification of diverse cognitive, robotic and VH capabilities provides an important first step towards fully adaptive and interactive VHs in Sigma.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Sigma is a nascent cognitive architecture/system that combines concepts from graphical models with traditional symbolic architectures. Here an initial Sigma-based virtual human (VH) is introduced that combines probabilistic reasoning, rule-based decision-making, Theory of Mind, Simultaneous Localization and Mapping and reinforcement learning in a unified manner. This non-modular unification of diverse cognitive, robotic and VH capabilities provides an important first step towards fully adaptive and interactive VHs in Sigma.
Rosenbloom, Paul S.; Gratch, Jonathan; Ustun, Volkan
Towards Emotion in Sigma: From Appraisal to Attention Proceedings Article
In: Proceedings of AGI 2015, pp. 142 – 151, Springer International Publishing, Berlin, Germany, 2015.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@inproceedings{rosenbloom_towards_2015,
title = {Towards Emotion in Sigma: From Appraisal to Attention},
author = {Paul S. Rosenbloom and Jonathan Gratch and Volkan Ustun},
url = {http://ict.usc.edu/pubs/Towards%20Emotion%20in%20Sigma%20-%20From%20Appraisal%20to%20Attention.pdf},
year = {2015},
date = {2015-07-01},
booktitle = {Proceedings of AGI 2015},
volume = {9205},
pages = {142 – 151},
publisher = {Springer International Publishing},
address = {Berlin, Germany},
abstract = {A first step is taken towards incorporating emotional processing into Sigma, a cognitive architecture that is grounded in graphical models, with the addition of appraisal variables for expectedness and desirability plus their initial implications for attention at two levels of the control hierarchy. The results leverage many of Sigma's existing capabilities but with a few key additions.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
A first step is taken towards incorporating emotional processing into Sigma, a cognitive architecture that is grounded in graphical models, with the addition of appraisal variables for expectedness and desirability plus their initial implications for attention at two levels of the control hierarchy. The results leverage many of Sigma's existing capabilities but with a few key additions.
Kommers, Cody; Ustun, Volkan; Demski, Abram; Rosenbloom, Paul
Hierarchical Reasoning with Distributed Vector Representations Proceedings Article
In: Proceedings of 37th Annual Conference of the Cognitive Science Society, Cognitive Science Society, Pasadena, CA, 2015.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@inproceedings{kommers_hierarchical_2015,
title = {Hierarchical Reasoning with Distributed Vector Representations},
author = {Cody Kommers and Volkan Ustun and Abram Demski and Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Hierarchical%20Reasoning%20with%20Distributed%20Vector%20Representations.pdf},
year = {2015},
date = {2015-07-01},
booktitle = {Proceedings of 37th Annual Conference of the Cognitive Science Society},
publisher = {Cognitive Science Society},
address = {Pasadena, CA},
abstract = {We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.
Rosenbloom, Paul
Supraarchitectural Capability Integration: From Soar to Sigma Proceedings Article
In: Proceedings of the 13th International Conference on Cognitive Modeling, 2015, Groningen, The Netherlands, 2015.
Abstract | Links | BibTeX | Tags: CogArch, UARC, Virtual Humans
@inproceedings{rosenbloom_supraarchitectural_2015,
title = {Supraarchitectural Capability Integration: From Soar to Sigma},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Supraarchitectural%20Capability%20Integration%20-%20From%20Soar%20to%20Sigma.pdf},
year = {2015},
date = {2015-04-01},
booktitle = {Proceedings of the 13th International Conference on Cognitive Modeling, 2015},
address = {Groningen, The Netherlands},
abstract = {Integration across capabilities, both architectural and supraarchitectural, is critical for cognitive architectures. Here we revisit a classic failure of supraarchitectural capability integration in Soar, involving data chunking, to understand better both its source and how it and related integration issues can be overcome via three general extensions in Sigma.},
keywords = {CogArch, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Integration across capabilities, both architectural and supraarchitectural, is critical for cognitive architectures. Here we revisit a classic failure of supraarchitectural capability integration in Soar, involving data chunking, to understand better both its source and how it and related integration issues can be overcome via three general extensions in Sigma.
Rosenbloom, Paul S.; Demski, Abram; Ustun, Volkan
Efficient message computation in Sigma’s graphical architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 11, pp. 1–9, 2015, ISSN: 2212683X.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC
@article{rosenbloom_efficient_2015,
title = {Efficient message computation in Sigma’s graphical architecture},
author = {Paul S. Rosenbloom and Abram Demski and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X14000723},
doi = {10.1016/j.bica.2014.11.009},
issn = {2212683X},
year = {2015},
date = {2015-01-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {11},
pages = {1–9},
abstract = {Human cognition runs at ∼50 ms per cognitive cycle, implying that any biologically inspired cognitive architecture that strives for real-time performance needs to be able to run at this speed. Sigma is a cognitive architecture built upon graphical models – a broadly applicable state-of-the-art formalism for implementing cognitive capabilities – that are solved via message passing (with complex messages based on n-dimensional piecewise-linear functions). Earlier work explored optimizations to Sigma that reduced by an order of magnitude the number of messages sent per cycle. Here, optimizations are introduced that reduce by an order of magnitude the average time required per message sent.},
keywords = {CogArch, Cognitive Architecture, UARC},
pubstate = {published},
tppubtype = {article}
}
Human cognition runs at ∼50 ms per cognitive cycle, implying that any biologically inspired cognitive architecture that strives for real-time performance needs to be able to run at this speed. Sigma is a cognitive architecture built upon graphical models – a broadly applicable state-of-the-art formalism for implementing cognitive capabilities – that are solved via message passing (with complex messages based on n-dimensional piecewise-linear functions). Earlier work explored optimizations to Sigma that reduced by an order of magnitude the number of messages sent per cycle. Here, optimizations are introduced that reduce by an order of magnitude the average time required per message sent.
2014
Joshi, Himanshu; Rosenbloom, Paul S.; Ustun, Volkan
Isolated word recognition in the Sigma cognitive architecture Journal Article
In: Biologically Inspired Cognitive Architectures, vol. 10, pp. 1–9, 2014, ISSN: 2212683X.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC
@article{joshi_isolated_2014,
title = {Isolated word recognition in the Sigma cognitive architecture},
author = {Himanshu Joshi and Paul S. Rosenbloom and Volkan Ustun},
url = {http://linkinghub.elsevier.com/retrieve/pii/S2212683X14000644},
doi = {10.1016/j.bica.2014.11.001},
issn = {2212683X},
year = {2014},
date = {2014-10-01},
journal = {Biologically Inspired Cognitive Architectures},
volume = {10},
pages = {1–9},
abstract = {Symbolic architectures are effective at complex cognitive reasoning, but typically are incapable of important forms of sub-cognitive processing – such as perception – without distinct modules connected to them via low-bandwidth interfaces. Neural architectures, in contrast, may be quite effective at the latter, but typically struggle with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. Here it is shown that Sigma is not only capable of performing a simple variant of speech recognition via the same knowledge structures and reasoning algorithm used for cognitive processing, but also of leveraging its existing knowledge templates and learning algorithm to acquire automatically most of the structures and parameters needed for this recognition activity.},
keywords = {CogArch, Cognitive Architecture, UARC},
pubstate = {published},
tppubtype = {article}
}
Symbolic architectures are effective at complex cognitive reasoning, but typically are incapable of important forms of sub-cognitive processing – such as perception – without distinct modules connected to them via low-bandwidth interfaces. Neural architectures, in contrast, may be quite effective at the latter, but typically struggle with the former. Sigma has been designed to leverage the state-of-the-art hybrid (discrete + continuous) mixed (symbolic + probabilistic) capability of graphical models to provide in a uniform non-modular fashion effective forms of, and integration across, both cognitive and sub-cognitive behavior. Here it is shown that Sigma is not only capable of performing a simple variant of speech recognition via the same knowledge structures and reasoning algorithm used for cognitive processing, but also of leveraging its existing knowledge templates and learning algorithm to acquire automatically most of the structures and parameters needed for this recognition activity.
Ustun, Volkan; Rosenbloom, Paul S.; Sagae, Kenji; Demski, Abram
Distributed Vector Representations of Words in the Sigma Cognitive Architecture Proceedings Article
In: Proceedings of the 7th Conference on Artificial General Intelligence 2014, Québec City, Canada, 2014.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{ustun_distributed_2014,
title = {Distributed Vector Representations of Words in the Sigma Cognitive Architecture},
author = {Volkan Ustun and Paul S. Rosenbloom and Kenji Sagae and Abram Demski},
url = {http://ict.usc.edu/pubs/Distributed%20Vector%20Representations%20of%20Words%20in%20the%20Sigma%20Cognitive%20Architecture.pdf},
year = {2014},
date = {2014-08-01},
booktitle = {Proceedings of the 7th Conference on Artificial General Intelligence 2014},
address = {Québec City, Canada},
abstract = {Recently reported results with distributed-vector word representations in natural language processing make them appealing for incorporation into a general cognitive architecture like Sigma. This paper describes a new algorithm for learning such word representations from large, shallow information resources, and how this algorithm can be implemented via small modifications to Sigma. The effectiveness and speed of the algorithm are evaluated via a comparison of an external simulation of it with state-of-the-art algorithms. The results from more limited experiments with Sigma are also promising, but more work is required for it to reach the effectiveness and speed of the simulation.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Recently reported results with distributed-vector word representations in natural language processing make them appealing for incorporation into a general cognitive architecture like Sigma. This paper describes a new algorithm for learning such word representations from large, shallow information resources, and how this algorithm can be implemented via small modifications to Sigma. The effectiveness and speed of the algorithm are evaluated via a comparison of an external simulation of it with state-of-the-art algorithms. The results from more limited experiments with Sigma are also promising, but more work is required for it to reach the effectiveness and speed of the simulation.
Rosenbloom, Paul S.
Ubiquity symposium: The science in computer science: the computing sciences and STEM education Journal Article
In: Ubiquity, no. March 2014, pp. 1–8, 2014, ISSN: 15302180.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture
@article{rosenbloom_ubiquity_2014,
title = {Ubiquity symposium: The science in computer science: the computing sciences and STEM education},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Ubiquity%20symposium%20-%20The%20science%20in%20computer%20science%20-%20the%20computing%20sciences%20and%20STEM%20education.pdf},
doi = {10.1145/2590528.2590530},
issn = {15302180},
year = {2014},
date = {2014-03-01},
journal = {Ubiquity},
number = {March 2014},
pages = {1–8},
abstract = {In this latest installment of "The Science in Computer Science," Prof. Paul Rosenbloom continues the discussion on whether or not computer science can be considered a "natural science." He argues not only is computing the basis for a true science, it is in fact an entire scientific domain.},
keywords = {CogArch, Cognitive Architecture},
pubstate = {published},
tppubtype = {article}
}
In this latest installment of "The Science in Computer Science," Prof. Paul Rosenbloom continues the discussion on whether or not computer science can be considered a "natural science." He argues not only is computing the basis for a true science, it is in fact an entire scientific domain.
2013
Pynadath, David V.; Rosenbloom, Paul; Marsella, Stacy C.; Li, Lingshan
Modeling Two-Player Games in the Sigma Graphical Cognitive Architecture Proceedings Article
In: Proceedings of the 6th Conference on Artificial General Intelligence, pp. 98–108, Beijing, China, 2013.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{pynadath_modeling_2013,
title = {Modeling Two-Player Games in the Sigma Graphical Cognitive Architecture},
author = {David V. Pynadath and Paul Rosenbloom and Stacy C. Marsella and Lingshan Li},
url = {http://ict.usc.edu/pubs/Modeling%20Two-Player%20Games%20in%20the%20Sigma%20Graphical%20Cognitive%20Architecture.pdf},
year = {2013},
date = {2013-07-01},
booktitle = {Proceedings of the 6th Conference on Artificial General Intelligence},
pages = {98–108},
address = {Beijing, China},
abstract = {Effective social interaction and, in particular, a Theory of Mind are critical components of human intelligence, allowing us to form beliefs about other people, generate expectations about their behavior, and use those expectations to inform our own decision-making. This article presents an investigation into methods for realizing Theory of Mind within Sigma, a graphical cognitive architecture. By extending the architecture to capture independent decisions and problem-solving for multiple agents, we implemented Sigma models of several canonical examples from game theory. We show that the resulting Sigma agents can capture the same behaviors prescribed by equilibrium solutions.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Effective social interaction and, in particular, a Theory of Mind are critical components of human intelligence, allowing us to form beliefs about other people, generate expectations about their behavior, and use those expectations to inform our own decision-making. This article presents an investigation into methods for realizing Theory of Mind within Sigma, a graphical cognitive architecture. By extending the architecture to capture independent decisions and problem-solving for multiple agents, we implemented Sigma models of several canonical examples from game theory. We show that the resulting Sigma agents can capture the same behaviors prescribed by equilibrium solutions.
Rosenbloom, Paul S.; Demski, Abram; Han, Teawon; Ustun, Volkan
Learning via Gradient Descent in Sigma Proceedings Article
In: International Conference on Cognitive Modeling, Ottawa, Canada, 2013.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{rosenbloom_learning_2013,
title = {Learning via Gradient Descent in Sigma},
author = {Paul S. Rosenbloom and Abram Demski and Teawon Han and Volkan Ustun},
url = {http://ict.usc.edu/pubs/Learning%20via%20Gradient%20Descent%20in%20Sigma.pdf},
year = {2013},
date = {2013-07-01},
booktitle = {International Conference on Cognitive Modeling},
address = {Ottawa, Canada},
abstract = {Integrating a gradient-descent learning mechanism at the core of the graphical models upon which the Sigma cognitive architecture/system is built yields learning behaviors that span important forms of both procedural learning (e.g., action and reinforcement learning) and declarative learning (e.g., supervised and unsupervised concept formation), plus several additional forms of learning (e.g., distribution tracking and map learning) relevant to cognitive systems/modeling. The core result presented here is this breadth of cognitive learning behaviors that is producible in this uniform manner.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Integrating a gradient-descent learning mechanism at the core of the graphical models upon which the Sigma cognitive architecture/system is built yields learning behaviors that span important forms of both procedural learning (e.g., action and reinforcement learning) and declarative learning (e.g., supervised and unsupervised concept formation), plus several additional forms of learning (e.g., distribution tracking and map learning) relevant to cognitive systems/modeling. The core result presented here is this breadth of cognitive learning behaviors that is producible in this uniform manner.
Rosenbloom, Paul S.
The Sigma cognitive architecture and system Journal Article
In: AISB Quarterly, pp. 4–13, 2013.
Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@article{rosenbloom_sigma_2013,
title = {The Sigma cognitive architecture and system},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/The%20Sigma%20cognitive%20architecture%20and%20system.pdf},
year = {2013},
date = {2013-05-01},
journal = {AISB Quarterly},
pages = {4–13},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
2012
Rosenbloom, Paul S.
Leveraging Computing Sciences in STEM Education Journal Article
In: ACM Ubiquity, 2012.
Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@article{rosenbloom_leveraging_2012,
title = {Leveraging Computing Sciences in STEM Education},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Leveraging%20Computing%20Sciences%20in%20STEM%20Education.pdf},
year = {2012},
date = {2012-12-01},
journal = {ACM Ubiquity},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
Demski, Abram
Logical Prior Probability Proceedings Article
In: Conference on Artificial General Intelligence, Oxford, UK, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{demski_logical_2012,
title = {Logical Prior Probability},
author = {Abram Demski},
url = {http://ict.usc.edu/pubs/Logical%20Prior%20Probability.pdf},
year = {2012},
date = {2012-12-01},
booktitle = {Conference on Artificial General Intelligence},
address = {Oxford, UK},
abstract = {A Bayesian prior over first-order theories is defined. It is shown that the prior can be approximated, and the relationship to previously studied priors is examined.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
A Bayesian prior over first-order theories is defined. It is shown that the prior can be approximated, and the relationship to previously studied priors is examined.
Rosenbloom, Paul
On Computing: The Fourth Great Scientific Domain Book
MIT Press, 2012, ISBN: 0-262-01832-2.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@book{rosenbloom_computing_2012,
title = {On Computing: The Fourth Great Scientific Domain},
author = {Paul Rosenbloom},
url = {http://www.amazon.com/On-Computing-Fourth-Scientific-Domain/dp/0262018322},
isbn = {0-262-01832-2},
year = {2012},
date = {2012-11-01},
publisher = {MIT Press},
abstract = {Computing isn't simply about hardware or software, or calculation or applications. Computing, writes Paul Rosenbloom, is an exciting and diverse, yet remarkably coherent, scientific enterprise that is highly multidisciplinary yet maintains a unique core of its own. In On Computing, Rosenbloom proposes that computing is a great scientific domain on a par with the physical, life, and social sciences. Rosenbloom introduces a relational approach for understanding computing, conceptualizing it in terms of forms of interaction and implementation, to reveal the hidden structures and connections among its disciplines. He argues for the continuing vitality of computing, surveying the leading edge in computing's combination with other domains, from biocomputing and brain-computer interfaces to crowdsourcing and virtual humans to robots and the intermingling of the real and the virtual. He explores forms of higher order coherence, or macrostructures, over complex computing topics and organizations, such as computing's role in the pursuit of science and the structure of academic computing. Finally, he examines the very notion of a great scientific domain in philosophical terms, honing his argument that computing should be considered the fourth great scientific domain. Rosenbloom's proposal may prove to be controversial, but the intent is to initiate a long overdue conversation about the nature and future of a field in search of its soul. Rosenbloom, a key architect of the founding of University of Southern California's Institute for Creative Technologies and former Deputy Director of USC's Information Sciences Institute, offers a broader perspective on what computing is and what it can becom},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {book}
}
Computing isn't simply about hardware or software, or calculation or applications. Computing, writes Paul Rosenbloom, is an exciting and diverse, yet remarkably coherent, scientific enterprise that is highly multidisciplinary yet maintains a unique core of its own. In On Computing, Rosenbloom proposes that computing is a great scientific domain on a par with the physical, life, and social sciences. Rosenbloom introduces a relational approach for understanding computing, conceptualizing it in terms of forms of interaction and implementation, to reveal the hidden structures and connections among its disciplines. He argues for the continuing vitality of computing, surveying the leading edge in computing's combination with other domains, from biocomputing and brain-computer interfaces to crowdsourcing and virtual humans to robots and the intermingling of the real and the virtual. He explores forms of higher order coherence, or macrostructures, over complex computing topics and organizations, such as computing's role in the pursuit of science and the structure of academic computing. Finally, he examines the very notion of a great scientific domain in philosophical terms, honing his argument that computing should be considered the fourth great scientific domain. Rosenbloom's proposal may prove to be controversial, but the intent is to initiate a long overdue conversation about the nature and future of a field in search of its soul. Rosenbloom, a key architect of the founding of University of Southern California's Institute for Creative Technologies and former Deputy Director of USC's Information Sciences Institute, offers a broader perspective on what computing is and what it can becom
Rosenbloom, Paul
Towards a Conceptual Framework for Digital Humanities Journal Article
In: Digital Humanities Quarterly, vol. 2, no. 6, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@article{rosenbloom_towards_2012-1,
title = {Towards a Conceptual Framework for Digital Humanities},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20a%20Conceptual%20Framework%20for%20Digital%20Humanities.pdf},
year = {2012},
date = {2012-10-01},
booktitle = {Proceedings of the 11th International Conference on Cognitive Modeling},
journal = {Digital Humanities Quarterly},
volume = {2},
number = {6},
address = {Berlin, Germany},
abstract = {The concept of a great scientific domain broadens what is normally considered to be within the purview of science while identifying four such domains – the physical, life, social and computing sciences – and suggesting that the humanities naturally fit within the sciences as part of an expanded social domain. The relational architecture that has been developed to aid in understanding disciplinary combinations across great scientific domains then guides an exploration of the structure and content of the digital humanities in terms of a space of relationships between computing and the humanities.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
The concept of a great scientific domain broadens what is normally considered to be within the purview of science while identifying four such domains – the physical, life, social and computing sciences – and suggesting that the humanities naturally fit within the sciences as part of an expanded social domain. The relational architecture that has been developed to aid in understanding disciplinary combinations across great scientific domains then guides an exploration of the structure and content of the digital humanities in terms of a space of relationships between computing and the humanities.
Rosenbloom, Paul S.
Extending Mental Imagery in Sigma Proceedings Article
In: Conference on Artificial General Intelligence, Oxford, UK, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{rosenbloom_extending_2012,
title = {Extending Mental Imagery in Sigma},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Extending%20Mental%20Imagery%20in%20Sigma.pdf},
year = {2012},
date = {2012-10-01},
booktitle = {Conference on Artificial General Intelligence},
address = {Oxford, UK},
abstract = {This article presents new results on implementing mental imagery within the Sigma cognitive architecture. Rather than amounting to a distinct module, mental imagery is based on the same primitive, hybrid mixed, architectural mechanisms as Sigma's other cognitive capabilities. The work here demonstrates the creation and modification of compound images, the transformation of individual objects within such images, and the extraction of derived information from these compositions.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
This article presents new results on implementing mental imagery within the Sigma cognitive architecture. Rather than amounting to a distinct module, mental imagery is based on the same primitive, hybrid mixed, architectural mechanisms as Sigma's other cognitive capabilities. The work here demonstrates the creation and modification of compound images, the transformation of individual objects within such images, and the extraction of derived information from these compositions.
Rosenbloom, Paul S.
Deconstructing Reinforcement Learning in Sigma Proceedings Article
In: Conference on Artificial General Intelligence, Oxford, UK, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, UARC, Virtual Humans
@inproceedings{rosenbloom_deconstructing_2012,
title = {Deconstructing Reinforcement Learning in Sigma},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Deconstructing%20Reinforcement%20Learning%20in%20Sigma.pdf},
year = {2012},
date = {2012-10-01},
booktitle = {Conference on Artificial General Intelligence},
address = {Oxford, UK},
abstract = {This article describes the development of reinforcement learning within the Sigma graphical cognitive architecture. Reinforcement learning has been deconstructed in terms of the interactions among more basic mechanisms and knowledge in Sigma, making it a derived capability rather than a de novo mechanism. Basic reinforcement learning – both model-based and model-free – are demonstrated, along with the intertwining of model learning.},
keywords = {CogArch, Cognitive Architecture, UARC, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
This article describes the development of reinforcement learning within the Sigma graphical cognitive architecture. Reinforcement learning has been deconstructed in terms of the interactions among more basic mechanisms and knowledge in Sigma, making it a derived capability rather than a de novo mechanism. Basic reinforcement learning – both model-based and model-free – are demonstrated, along with the intertwining of model learning.
Rosenbloom, Paul
Towards a 50 msec Cognitive Cycle in a Graphical Architecture Proceedings Article
In: Proceedings of the 11th International Conference on Cognitive Modeling, Berlin, Germany, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_towards_2012,
title = {Towards a 50 msec Cognitive Cycle in a Graphical Architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20a%2050%20msec%20Cognitive%20Cycle%20in%20a%20Graphical%20Architecture.pdf},
year = {2012},
date = {2012-04-01},
booktitle = {Proceedings of the 11th International Conference on Cognitive Modeling},
volume = {2},
number = {6},
address = {Berlin, Germany},
abstract = {Achieving a 50 msec cognitive cycle in any sufficiently sophisticated cognitive architecture can be a significant challenge. Here an investigation is begun into how to do this within a recently developed graphical architecture that is based on factor graphs (with the summary product algorithm) and piecewise continuous functions. Results are presented from three optimizations that leverage the structure of factor graphs to reduce the number of message cycles required per cognitive cycle.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Achieving a 50 msec cognitive cycle in any sufficiently sophisticated cognitive architecture can be a significant challenge. Here an investigation is begun into how to do this within a recently developed graphical architecture that is based on factor graphs (with the summary product algorithm) and piecewise continuous functions. Results are presented from three optimizations that leverage the structure of factor graphs to reduce the number of message cycles required per cognitive cycle.
Rosenbloom, Paul
Towards functionally elegant, grand unified architectures, part of Accelerating the Evolution of Cognitive Architecture Proceedings Article
In: Behavior Representation in Modeling & Simulation (BRIMS) Conference, 2012.
Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_towards_2012-2,
title = {Towards functionally elegant, grand unified architectures, part of Accelerating the Evolution of Cognitive Architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20functionally%20elegant%20grand%20unified%20architectures.pdf},
year = {2012},
date = {2012-03-01},
booktitle = {Behavior Representation in Modeling & Simulation (BRIMS) Conference},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Rosenbloom, Paul
Graphical Models for Integrated Intelligent Robot Architectures Proceedings Article
In: Proceedings of the AAAI Spring Symposium on Designing Intelligent Robots: Reintegrating AI, Stanford University, 2012.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_graphical_2012,
title = {Graphical Models for Integrated Intelligent Robot Architectures},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Graphical%20Models%20for%20Integrated%20Intelligent%20Robot%20Architectures.pdf},
year = {2012},
date = {2012-03-01},
booktitle = {Proceedings of the AAAI Spring Symposium on Designing Intelligent Robots: Reintegrating AI},
address = {Stanford University},
abstract = {The theoretically elegant yet broadly functional capability of graphical models shows intriguing potential to span in a uniform manner perception, cognition and action; and thus to ultimately yield simpler yet more powerful integrated architectures for intelligent robots and other comparable systems. This position paper explores this potential, with initial support from an effort underway to develop a graphical architecture that is based on factor graphs (with piecewise continuous functions).},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
The theoretically elegant yet broadly functional capability of graphical models shows intriguing potential to span in a uniform manner perception, cognition and action; and thus to ultimately yield simpler yet more powerful integrated architectures for intelligent robots and other comparable systems. This position paper explores this potential, with initial support from an effort underway to develop a graphical architecture that is based on factor graphs (with piecewise continuous functions).
2011
Rosenbloom, Paul
Bridging dichotomies in cognitive architectures for virtual humans Proceedings Article
In: Proceedings of the AAAI Fall Symposium on Advances in Cognitive Systems, Arlington, VA, 2011.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_bridging_2011,
title = {Bridging dichotomies in cognitive architectures for virtual humans},
author = {Paul Rosenbloom},
url = {http://www.ict.usc.edu/pubs/Bridging%20Dichotomies%20in%20Cognitive%20Architectures%20for%20Virtual%20Humans.pdf},
year = {2011},
date = {2011-11-01},
booktitle = {Proceedings of the AAAI Fall Symposium on Advances in Cognitive Systems},
address = {Arlington, VA},
abstract = {Desiderata for cognitive architectures that are to support the extent of human-level intelligence required in virtual humans imply the need to bridge a range of dichotomies faced by such architectures. The focus here is first on two general approaches to building such bridges – addition and reduction – and then on a pair of general tools – graphical models and piecewise continuous functions – that exploit the second approach towards developing such an architecture. Evaluation is in terms of the architecture's demonstrated ability and future potential for bridging the dichotomies.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Desiderata for cognitive architectures that are to support the extent of human-level intelligence required in virtual humans imply the need to bridge a range of dichotomies faced by such architectures. The focus here is first on two general approaches to building such bridges – addition and reduction – and then on a pair of general tools – graphical models and piecewise continuous functions – that exploit the second approach towards developing such an architecture. Evaluation is in terms of the architecture's demonstrated ability and future potential for bridging the dichotomies.
Chen, Junda; Demski, Abram; Han, Teawon; Morency, Louis-Philippe; Pynadath, David V.; Rafidi, Nicole; Rosenbloom, Paul
Fusing symbolic and decision-theoretic problem solving + perception in a graphical cognitive architecture Proceedings Article
In: Proceedings of the 2nd International Conference on Biologically Inspired Cognitive Architecture (BICA), Arlington, VA, 2011.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Social Simulation, Virtual Humans
@inproceedings{chen_fusing_2011,
title = {Fusing symbolic and decision-theoretic problem solving + perception in a graphical cognitive architecture},
author = {Junda Chen and Abram Demski and Teawon Han and Louis-Philippe Morency and David V. Pynadath and Nicole Rafidi and Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Fusing%20symbolic%20and%20decision-theoretic%20problem%20solving%20+%20perception%20in%20a%20graphical%20cognitive%20architecture.pdf},
year = {2011},
date = {2011-11-01},
booktitle = {Proceedings of the 2nd International Conference on Biologically Inspired Cognitive Architecture (BICA)},
address = {Arlington, VA},
abstract = {A step is taken towards fusing symbolic and decision-theoretic problem solving in a cognitive architecture by implementing the latter in an architecture within which the former has already been demonstrated. The graphical models upon which the architecture is based enable a uniform implementation of both varieties of problem solving. They also enable a uniform combination with forms of decision-relevant perception, highlighting a potential path towards a tight coupling between central cognition and peripheral perception.},
keywords = {CogArch, Cognitive Architecture, Social Simulation, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
A step is taken towards fusing symbolic and decision-theoretic problem solving in a cognitive architecture by implementing the latter in an architecture within which the former has already been demonstrated. The graphical models upon which the architecture is based enable a uniform implementation of both varieties of problem solving. They also enable a uniform combination with forms of decision-relevant perception, highlighting a potential path towards a tight coupling between central cognition and peripheral perception.
Rosenbloom, Paul
Mental imagery in a graphical cognitive architecture Proceedings Article
In: Proceedings of the 2nd International Conference on Biologically Inspired Cognitive Architecture (BICA), Arlington, VA, 2011.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_mental_2011,
title = {Mental imagery in a graphical cognitive architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Mental%20imagery%20in%20a%20graphical%20cognitive%20architecture.pdf},
year = {2011},
date = {2011-11-01},
booktitle = {Proceedings of the 2nd International Conference on Biologically Inspired Cognitive Architecture (BICA)},
address = {Arlington, VA},
abstract = {Can mental imagery be incorporated uniformly into a cognitive architecture by leveraging the mixed (relational and probabilistic) hybrid (discrete and continuous) processing supported by factor graphs? This article takes an initial step towards answering this question via an implementation in a graphical cognitive architecture of a version of the Eight Puzzle based on a hybrid function representation and a factor node optimized for object translation.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Can mental imagery be incorporated uniformly into a cognitive architecture by leveraging the mixed (relational and probabilistic) hybrid (discrete and continuous) processing supported by factor graphs? This article takes an initial step towards answering this question via an implementation in a graphical cognitive architecture of a version of the Eight Puzzle based on a hybrid function representation and a factor node optimized for object translation.
Rosenbloom, Paul
From memory to problem solving: Mechanism reuse in a graphical cognitive architecture Proceedings Article
In: Proceedings of the 4th Conference on Artificial General Intelligence, Mountain View, CA, 2011.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_memory_2011,
title = {From memory to problem solving: Mechanism reuse in a graphical cognitive architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/From%20memory%20to%20problem%20solving-%20Mechanism%20reuse%20in%20a%20graphical%20cognitive%20architecture.pdf},
year = {2011},
date = {2011-08-01},
booktitle = {Proceedings of the 4th Conference on Artificial General Intelligence},
address = {Mountain View, CA},
abstract = {This article describes the extension of a memory architecture that is implemented via graphical models to include core aspects of problem solving. By extensive reuse of the general graphical mechanisms originally developed to support memory, this demonstrates how a theoretically elegant implementation level can enable increasingly broad architectures without compromising overall simplicity and uniformity. In the process, it bolsters the potential of such an approach for developing the more complete architectures that will ultimately be necessary to support autonomous general intelligence.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
This article describes the extension of a memory architecture that is implemented via graphical models to include core aspects of problem solving. By extensive reuse of the general graphical mechanisms originally developed to support memory, this demonstrates how a theoretically elegant implementation level can enable increasingly broad architectures without compromising overall simplicity and uniformity. In the process, it bolsters the potential of such an approach for developing the more complete architectures that will ultimately be necessary to support autonomous general intelligence.
Rosenbloom, Paul
Rethinking Cognitive Architecture via Graphical Models Journal Article
In: Cognitive Systems Research, 2011.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@article{rosenbloom_rethinking_2011,
title = {Rethinking Cognitive Architecture via Graphical Models},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Rethinking%20Cognitive%20Architecture%20via%20Graphical%20Models.pdf},
year = {2011},
date = {2011-03-01},
journal = {Cognitive Systems Research},
abstract = {Cognitive architectures need to resolve the diversity dilemma – i.e., to blend diversity and simplicity – in order to couple functionality and efficiency with integrability, extensibility and maintainability. Building diverse architectures upon a uniform implementation level of graphical models is an intriguing approach because of the homogeneous manner in which such models produce state-of-the-art algorithms spanning symbol, probability and signal processing. To explore this approach a hybrid (discrete and continuous) mixed (Boolean and Bayesian) version of the Soar architecture is being implemented via graphical models. Initial steps reported here begin to show the potential of such an approach for cognitive architecture.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
Cognitive architectures need to resolve the diversity dilemma – i.e., to blend diversity and simplicity – in order to couple functionality and efficiency with integrability, extensibility and maintainability. Building diverse architectures upon a uniform implementation level of graphical models is an intriguing approach because of the homogeneous manner in which such models produce state-of-the-art algorithms spanning symbol, probability and signal processing. To explore this approach a hybrid (discrete and continuous) mixed (Boolean and Bayesian) version of the Soar architecture is being implemented via graphical models. Initial steps reported here begin to show the potential of such an approach for cognitive architecture.
2010
Rosenbloom, Paul
Computing and computation Journal Article
In: The Computer Journal, vol. 55, no. 7, pp. 820–824, 2010.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@article{rosenbloom_computing_2010,
title = {Computing and computation},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Computing%20and%20computation.pdf},
doi = {10.1093/comjnl/bxs070},
year = {2010},
date = {2010-12-01},
journal = {The Computer Journal},
volume = {55},
number = {7},
pages = {820–824},
abstract = {In this essay we claim that computing is the fourth great scientific domain, on par with the physical, life, and social sciences.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
In this essay we claim that computing is the fourth great scientific domain, on par with the physical, life, and social sciences.
Rosenbloom, Paul
Implementing First-Order Variables in a Graphical Cognitive Architecture Proceedings Article
In: Proceedings of the First International Conference on Biologically Inspired Cognitive Architectures, IOS Press, Washington, DC, 2010.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_implementing_2010,
title = {Implementing First-Order Variables in a Graphical Cognitive Architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Implementing%20First-Order%20Variables%20in%20a%20Graphical%20Cognitive%20Architecture.pdf},
year = {2010},
date = {2010-11-01},
booktitle = {Proceedings of the First International Conference on Biologically Inspired Cognitive Architectures},
publisher = {IOS Press},
address = {Washington, DC},
abstract = {Graphical cognitive architectures implement their functionality through localized message passing among computationally limited nodes. First-order variables – particularly universally quantified ones – while critical for some potential architectural mechanisms, can be quite difficult to implement in such architectures. A new implementation strategy based on message decomposition in graphical models is presented that yields tractability while preserving key symmetries in the graphs concerning how quantified variables are represented and how symbols, probabilities and signals are processed.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Graphical cognitive architectures implement their functionality through localized message passing among computationally limited nodes. First-order variables – particularly universally quantified ones – while critical for some potential architectural mechanisms, can be quite difficult to implement in such architectures. A new implementation strategy based on message decomposition in graphical models is presented that yields tractability while preserving key symmetries in the graphs concerning how quantified variables are represented and how symbols, probabilities and signals are processed.
Rosenbloom, Paul
Combining Procedural and Declarative Knowledge in a Graphical Architecture Proceedings Article
In: Proceedings of the 10th International Conference on Cognitive Modeling, Philadelphia, PA, 2010.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans, Virtual Worlds
@inproceedings{rosenbloom_combining_2010,
title = {Combining Procedural and Declarative Knowledge in a Graphical Architecture},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Combining%20Procedural%20and%20Declarative%20Knowledge%20in%20a%20Graphical%20Architecture.pdf},
year = {2010},
date = {2010-08-01},
booktitle = {Proceedings of the 10th International Conference on Cognitive Modeling},
address = {Philadelphia, PA},
abstract = {A prototypical cognitive architecture defines a memory architecture embodying forms of both procedural and declarative memory, plus their interaction. Reengineering such a dual architecture on a common foundation of graphical models enables a better understanding of both the substantial commonalities between procedural and declarative memory and the subtle differences that endow each with its own special character. It also opens the way towards blended capabilities that go beyond existing architectural memories.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans, Virtual Worlds},
pubstate = {published},
tppubtype = {inproceedings}
}
A prototypical cognitive architecture defines a memory architecture embodying forms of both procedural and declarative memory, plus their interaction. Reengineering such a dual architecture on a common foundation of graphical models enables a better understanding of both the substantial commonalities between procedural and declarative memory and the subtle differences that endow each with its own special character. It also opens the way towards blended capabilities that go beyond existing architectural memories.
Rosenbloom, Paul S.
Speculations on Leveraging Graphical Models for Architectural Integration of Visual Representation and Reasoning Proceedings Article
In: Proceedings of the AAAI-10 Workshop on Visual Representations and Reasoning, Atlanta, GA, 2010.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_speculations_2010,
title = {Speculations on Leveraging Graphical Models for Architectural Integration of Visual Representation and Reasoning},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Speculations%20on%20Leveraging%20Graphical%20Models%20for%20Architectural%20Integration%20of%20Visual%20Representation%20and%20Reasoning.pdf},
year = {2010},
date = {2010-07-01},
booktitle = {Proceedings of the AAAI-10 Workshop on Visual Representations and Reasoning},
address = {Atlanta, GA},
abstract = {Speculations on Leveraging Graphical Models for Architectural Integration of Visual Representation and Reasoning},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Speculations on Leveraging Graphical Models for Architectural Integration of Visual Representation and Reasoning
Rosenbloom, Paul S.
An Architectural Approach to Statistical Relational AI Proceedings Article
In: Proceedings of the AAAI-10 Workshop on Statistical Relational AI, Atlanta, GA, 2010.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_architectural_2010,
title = {An Architectural Approach to Statistical Relational AI},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/An%20Architectural%20Approach%20to%20Statistical%20Relational%20AI.pdf},
year = {2010},
date = {2010-01-01},
booktitle = {Proceedings of the AAAI-10 Workshop on Statistical Relational AI},
address = {Atlanta, GA},
abstract = {The architectural approach to AI focuses on the fixed structure underlying intelligence. Applying it to statistical relational AI could stimulate investigations of statistical relational approaches across AI, encourage understanding of commonalities and compatibilities across this range, and yield new architectures significantly beyond today's best.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
The architectural approach to AI focuses on the fixed structure underlying intelligence. Applying it to statistical relational AI could stimulate investigations of statistical relational approaches across AI, encourage understanding of commonalities and compatibilities across this range, and yield new architectures significantly beyond today's best.
2009
Rosenbloom, Paul S.
Towards Uniform Implementation of Architectural Diversity Proceedings Article
In: Proceedings of the AAAI Fall Symposium on Multi-Representational Architectures for Human-Level Intelligence, 2009.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_towards_2009-1,
title = {Towards Uniform Implementation of Architectural Diversity},
author = {Paul S. Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20Uniform%20Implementation%20of%20Architectural%20Diversity.pdf},
year = {2009},
date = {2009-11-01},
booktitle = {Proceedings of the AAAI Fall Symposium on Multi-Representational Architectures for Human-Level Intelligence},
abstract = {Multi-representational architectures exploit diversity to yield the breadth of capabilities required for intelligent behavior in the world, but in so doing can sacrifice too much of the complementary benefits of architectural uniformity. The proposal here is to couple the benefits of diversity and uniformity through establishment of a uniform graph-based implementation level for diverse architectures.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Multi-representational architectures exploit diversity to yield the breadth of capabilities required for intelligent behavior in the world, but in so doing can sacrifice too much of the complementary benefits of architectural uniformity. The proposal here is to couple the benefits of diversity and uniformity through establishment of a uniform graph-based implementation level for diverse architectures.
Rosenbloom, Paul
Towards a New Cognitive Hourglass: Uniform Implementation of Cognitive Architecture via Factor Graphs Proceedings Article
In: Proceedings of the 9th International Conference on Cognitive Modeling (ICCM 2009), Manchester, UK, 2009.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_towards_2009,
title = {Towards a New Cognitive Hourglass: Uniform Implementation of Cognitive Architecture via Factor Graphs},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Towards%20a%20New%20Cognitive%20Hourglass.pdf},
year = {2009},
date = {2009-07-01},
booktitle = {Proceedings of the 9th International Conference on Cognitive Modeling (ICCM 2009)},
address = {Manchester, UK},
abstract = {As cognitive architectures become ever more ambitious in the range of phenomena they are to assist in producing and modeling, there is increasing pressure for diversity in the mechanisms they embody. Yet uniformity remains critical for both elegance and extensibility. Here, the search for uniformity is continued, but shifted downwards in the cognitive hierarchy to the implementation level. Factor graphs are explored as a promising core, with initial steps towards a reimplementation of Soar. The ultimate aim is a uniform implementation level for cognitive architectures affording both heightened elegance and expanded coverage.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
As cognitive architectures become ever more ambitious in the range of phenomena they are to assist in producing and modeling, there is increasing pressure for diversity in the mechanisms they embody. Yet uniformity remains critical for both elegance and extensibility. Here, the search for uniformity is continued, but shifted downwards in the cognitive hierarchy to the implementation level. Factor graphs are explored as a promising core, with initial steps towards a reimplementation of Soar. The ultimate aim is a uniform implementation level for cognitive architectures affording both heightened elegance and expanded coverage.
Rosenbloom, Paul
A Graphical Rethinking of the Cognitive Inner Loop Proceedings Article
In: Proceedings of the The First International Workshop on Graphical Representations for Knowledge Representation and Reasoning, 2009.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@inproceedings{rosenbloom_graphical_2009,
title = {A Graphical Rethinking of the Cognitive Inner Loop},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/A%20Graphical%20Rethinking%20of%20the%20Cognitive%20Inner%20Loop.pdf},
year = {2009},
date = {2009-01-01},
booktitle = {Proceedings of the The First International Workshop on Graphical Representations for Knowledge Representation and Reasoning},
abstract = {Explorations of graphical representation and rea- soning have yielded intriguing results spanning symbol, probability and signal processing. Here we explore an integrative application of graphs, as a path towards cognitive architectures of increased elegance, functionality, and extensibility. The spe- cific focus is on steps towards a graphical reim- plementation and extension of the cognitive inner loop within the Soar architecture. Alchemy, an im- plementation of Markov logic, is used for initial experiments, yielding insights into what will ulti- mately be required for full graphical implementa- tions of enhanced cognitive inner loops.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
Explorations of graphical representation and rea- soning have yielded intriguing results spanning symbol, probability and signal processing. Here we explore an integrative application of graphs, as a path towards cognitive architectures of increased elegance, functionality, and extensibility. The spe- cific focus is on steps towards a graphical reim- plementation and extension of the cognitive inner loop within the Soar architecture. Alchemy, an im- plementation of Markov logic, is used for initial experiments, yielding insights into what will ulti- mately be required for full graphical implementa- tions of enhanced cognitive inner loops.
2006
Rosenbloom, Paul
A Cognitive Odyssey: From the Power Law of Practice to a General Learning Mechanism and Beyond Journal Article
In: Tutorials in Quantitative Methods for Psychology, vol. 2, no. 2, pp. 43–51, 2006.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Virtual Humans
@article{rosenbloom_cognitive_2006,
title = {A Cognitive Odyssey: From the Power Law of Practice to a General Learning Mechanism and Beyond},
author = {Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/A%20Cognitive%20Odyssey-%20From%20the%20Power%20Law%20of%20Practice%20to%20a%20General%20Learning%20Mechanism%20and%20Beyond.pdf},
year = {2006},
date = {2006-01-01},
journal = {Tutorials in Quantitative Methods for Psychology},
volume = {2},
number = {2},
pages = {43–51},
abstract = {This article traces a line of research that began with the establishment of a pervasive regularity in human performance – the Power Law of Practice – and proceeded through several decades' worth of investigations that this opened up into learning and cognitive architecture. The results touch on both cognitive psychology and artificial intelligence, and more specifically on the possibily of building general learning mechanisms/systems. It is a story whose final chapter is still to be written.},
keywords = {CogArch, Cognitive Architecture, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
This article traces a line of research that began with the establishment of a pervasive regularity in human performance – the Power Law of Practice – and proceeded through several decades' worth of investigations that this opened up into learning and cognitive architecture. The results touch on both cognitive psychology and artificial intelligence, and more specifically on the possibily of building general learning mechanisms/systems. It is a story whose final chapter is still to be written.
2000
Hill, Randall W.; Gratch, Jonathan; Rosenbloom, Paul
Flexible Group Behavior: Virtual Commanders for Synthetic Battlespaces Proceedings Article
In: Proceedings of the 4th International Conference on Autonomous Agents, Barcelona, Spain, 2000.
Abstract | Links | BibTeX | Tags: CogArch, Cognitive Architecture, Social Simulation, Virtual Humans
@inproceedings{hill_flexible_2000,
title = {Flexible Group Behavior: Virtual Commanders for Synthetic Battlespaces},
author = {Randall W. Hill and Jonathan Gratch and Paul Rosenbloom},
url = {http://ict.usc.edu/pubs/Flexible%20Group%20Behavior-%20Virtual%20Commanders%20for%20Synthetic%20Battlespaces.pdf},
year = {2000},
date = {2000-06-01},
booktitle = {Proceedings of the 4th International Conference on Autonomous Agents},
address = {Barcelona, Spain},
abstract = {This paper describes a project to develop autonomous commander agents for synthetic battlespaces. The commander agents plan missions, monitor their execution, and replan when necessary. To reason about the social aspects of group behavior, the commanders take various social stances that enable them to collaborate with friends, exercise or defer to authority, and thwart their foes. The purpose of this paper is to describe these capabilities and how they came to be through a series of lessons learned while developing autonomous agents for this domain.},
keywords = {CogArch, Cognitive Architecture, Social Simulation, Virtual Humans},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes a project to develop autonomous commander agents for synthetic battlespaces. The commander agents plan missions, monitor their execution, and replan when necessary. To reason about the social aspects of group behavior, the commanders take various social stances that enable them to collaborate with friends, exercise or defer to authority, and thwart their foes. The purpose of this paper is to describe these capabilities and how they came to be through a series of lessons learned while developing autonomous agents for this domain.