Publications
Search
501.
Lane, H. Chad; Core, Mark G.; Goldberg, Benjamin S.
Lowering the Technical Skill Requirements for Building Intelligent Tutors: A Review of Authoring Tools Book Section
In: Design Recommendations for Intelligent Tutoring Systems, vol. 3, pp. 303 – 318, U.S. Army Research Laboratory, 2015.
@incollection{lane_lowering_2015,
title = {Lowering the Technical Skill Requirements for Building Intelligent Tutors: A Review of Authoring Tools},
author = {H. Chad Lane and Mark G. Core and Benjamin S. Goldberg},
url = {http://ict.usc.edu/pubs/Lowering%20the%20Technical%20Skill%20Requirements%20for%20Building%20Intelligent%20Tutors-A%20Review%20of%20Authoring%20Tools.pdf},
year = {2015},
date = {2015-06-01},
booktitle = {Design Recommendations for Intelligent Tutoring Systems},
volume = {3},
pages = {303 – 318},
publisher = {U.S. Army Research Laboratory},
abstract = {In this chapter, we focus on intelligent tutoring systems (ITSs), an instance of educational technology that is often criticized for not reaching its full potential (Nye, 2013). Researchers have debated why, given such strong empirical evidence in their favor (Anderson, Corbett, Koedinger & Pelletier, 1995; D’Mello & Graesser, 2012; VanLehn et al., 2005; Woolf, 2009), intelligent tutors are not in every classroom, on every device, providing educators with fine-grained assessment information about their students. Although many factors contribute to a lack of adoption (Nye, 2014), one widely agreed upon reason behind slow adoption and poor scalability of ITSs is that the engineering demands are simply too great. This is no surprise given that the effectiveness of ITSs is often attributable to the use of rich knowledge representations and cognitively plausible models of domain knowledge (Mark & Greer, 1995; Valerie J. Shute & Psotka, 1996; VanLehn, 2006; Woolf, 2009), which are inherently burdensome to build. To put it another way: the features that tend to make ITSs effective are also the hardest to build. The heavy reliance on cognitive scientists and artificial intelligence (AI) software engineers seems to be a bottleneck.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
In this chapter, we focus on intelligent tutoring systems (ITSs), an instance of educational technology that is often criticized for not reaching its full potential (Nye, 2013). Researchers have debated why, given such strong empirical evidence in their favor (Anderson, Corbett, Koedinger & Pelletier, 1995; D’Mello & Graesser, 2012; VanLehn et al., 2005; Woolf, 2009), intelligent tutors are not in every classroom, on every device, providing educators with fine-grained assessment information about their students. Although many factors contribute to a lack of adoption (Nye, 2014), one widely agreed upon reason behind slow adoption and poor scalability of ITSs is that the engineering demands are simply too great. This is no surprise given that the effectiveness of ITSs is often attributable to the use of rich knowledge representations and cognitively plausible models of domain knowledge (Mark & Greer, 1995; Valerie J. Shute & Psotka, 1996; VanLehn, 2006; Woolf, 2009), which are inherently burdensome to build. To put it another way: the features that tend to make ITSs effective are also the hardest to build. The heavy reliance on cognitive scientists and artificial intelligence (AI) software engineers seems to be a bottleneck.
502.
Chatterjee, Moitreya; Leuski, Anton
CRMActive: An Active Learning Based Approach for Effective Video Annotation and Retrieval Proceedings Article
In: Proceedings of ACM International Conference on Multimedia Retrieval (ICMR), pp. 535–538, ACM, Shanghai, China, 2015.
@inproceedings{chatterjee_crmactive_2015,
title = {CRMActive: An Active Learning Based Approach for Effective Video Annotation and Retrieval},
author = {Moitreya Chatterjee and Anton Leuski},
url = {http://ict.usc.edu/pubs/CRMActive%20-%20An%20Active%20Learning%20Based%20Approach%20for%20Effective%20Video%20Annotation%20and%20Retrieval.pdf},
doi = {10.1145/2671188.2749342},
year = {2015},
date = {2015-06-01},
booktitle = {Proceedings of ACM International Conference on Multimedia Retrieval (ICMR)},
pages = {535–538},
publisher = {ACM},
address = {Shanghai, China},
abstract = {Conventional multimedia annotation/retrieval systems such as Normalized Continuous Relevance Model (NormCRM) [7]require a fully labeled training data for a good performance. Active Learning, by determining an order for labeling the training data, allows for a good performance even before the training data is fully annotated. In this work we propose an active learning algorithm, which combines a novel measure of sample uncertainty with a novel clustering-based approach for determining sample density and diversity and integrate it with NormCRM. The clusters are also iteratively re⬚ned to ensure both feature and label-level agreement among samples. We show that our approach outperforms multiple baselines both on a new, open dataset and on the popular TRECVID corpus at both the tasks of annotation and text-based retrieval of videos.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Conventional multimedia annotation/retrieval systems such as Normalized Continuous Relevance Model (NormCRM) [7]require a fully labeled training data for a good performance. Active Learning, by determining an order for labeling the training data, allows for a good performance even before the training data is fully annotated. In this work we propose an active learning algorithm, which combines a novel measure of sample uncertainty with a novel clustering-based approach for determining sample density and diversity and integrate it with NormCRM. The clusters are also iteratively re⬚ned to ensure both feature and label-level agreement among samples. We show that our approach outperforms multiple baselines both on a new, open dataset and on the popular TRECVID corpus at both the tasks of annotation and text-based retrieval of videos.
503.
Garten, Justin; Sagae, Kenji; Ustun, Volkan; Dehghani, Morteza
Combining Distributed Vector Representations for Words Proceedings Article
In: Proceedings of NAACL-HLT 2015, pp. 95–101, Association for Computational Linguistics, Denver, Colorado, 2015.
@inproceedings{garten_combining_2015,
title = {Combining Distributed Vector Representations for Words},
author = {Justin Garten and Kenji Sagae and Volkan Ustun and Morteza Dehghani},
url = {http://ict.usc.edu/pubs/Combining%20Distributed%20Vector%20Representations%20for%20Words.pdf},
year = {2015},
date = {2015-06-01},
booktitle = {Proceedings of NAACL-HLT 2015},
pages = {95–101},
publisher = {Association for Computational Linguistics},
address = {Denver, Colorado},
abstract = {Recent interest in distributed vector representations for words has resulted in an increased diversity of approaches, each with strengths and weaknesses. We demonstrate how diverse vector representations may be inexpensively composed into hybrid representations, effectively leveraging strengths of individual components, as evidenced by substantial improvements on a standard word analogy task. We further compare these results over different sizes of training sets and find these advantages are more pronounced when training data is limited. Finally, we explore the relative impacts of the differences in the learning methods themselves and the size of the contexts they access.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Recent interest in distributed vector representations for words has resulted in an increased diversity of approaches, each with strengths and weaknesses. We demonstrate how diverse vector representations may be inexpensively composed into hybrid representations, effectively leveraging strengths of individual components, as evidenced by substantial improvements on a standard word analogy task. We further compare these results over different sizes of training sets and find these advantages are more pronounced when training data is limited. Finally, we explore the relative impacts of the differences in the learning methods themselves and the size of the contexts they access.
504.
Artstein, Ron; Leuski, Anton; Maio, Heather; Mor-Barak, Tomer; Gordon, Carla; Traum, David
How Many Utterances Are Needed to Support Time-Offset Interaction? Proceedings Article
In: Proceedings of FLAIRS 28, pp. 144–149, AAAI Press, Hollywood, FL, 2015, ISBN: 978-1-57735-730-8.
@inproceedings{artstein_how_2015,
title = {How Many Utterances Are Needed to Support Time-Offset Interaction?},
author = {Ron Artstein and Anton Leuski and Heather Maio and Tomer Mor-Barak and Carla Gordon and David Traum},
url = {http://www.aaai.org/ocs/index.php/FLAIRS/FLAIRS15/paper/view/10442},
isbn = {978-1-57735-730-8},
year = {2015},
date = {2015-05-01},
booktitle = {Proceedings of FLAIRS 28},
pages = {144–149},
publisher = {AAAI Press},
address = {Hollywood, FL},
abstract = {A set of several hundred recorded statements by a single speaker is sufficient to address unrestricted questions and sustain short conversations on a circumscribed topic. Statements were recorded by Pinchas Gutter, a Holocaust survivor, talking about his personal experiences before, during and after the Holocaust. These statements were delivered to participants in conversation, using a “Wizard of Oz” system, where live operators select an appropriate reaction to each user utterance in real time. Even though participants were completely unconstrained in the questions they could ask, the recorded statements were able to directly address at least 58% of user questions. The unanswered questions were then analyzed to identify gaps, and additional statements were recorded to fill the gaps. The statements will be put in an automated system using existing language understanding technology, to create the first full working system of time-offset interaction, allowing a live conversation with a real human who is not present for the conversation in real time.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A set of several hundred recorded statements by a single speaker is sufficient to address unrestricted questions and sustain short conversations on a circumscribed topic. Statements were recorded by Pinchas Gutter, a Holocaust survivor, talking about his personal experiences before, during and after the Holocaust. These statements were delivered to participants in conversation, using a “Wizard of Oz” system, where live operators select an appropriate reaction to each user utterance in real time. Even though participants were completely unconstrained in the questions they could ask, the recorded statements were able to directly address at least 58% of user questions. The unanswered questions were then analyzed to identify gaps, and additional statements were recorded to fill the gaps. The statements will be put in an automated system using existing language understanding technology, to create the first full working system of time-offset interaction, allowing a live conversation with a real human who is not present for the conversation in real time.
505.
Mell, Johnathan; Lucas, Gale; Gratch, Jonathan
An Effective Conversation Tactic for Creating Value over Repeated Negotiations Proceedings Article
In: Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems, pp. 1567–1576, International Foundation for Autonomous Agents and Multiagent Systems, Istanbul, Turkey, 2015.
@inproceedings{mell_effective_2015,
title = {An Effective Conversation Tactic for Creating Value over Repeated Negotiations},
author = {Johnathan Mell and Gale Lucas and Jonathan Gratch},
url = {http://ict.usc.edu/pubs/An%20Effective%20Conversation%20Tactic%20for%20Creating%20Value%20over%20Repeated%20Negotiations.pdf},
year = {2015},
date = {2015-05-01},
booktitle = {Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems},
pages = {1567–1576},
publisher = {International Foundation for Autonomous Agents and Multiagent Systems},
address = {Istanbul, Turkey},
abstract = {Automated negotiation research focuses on getting the most value from a single negotiation, yet real-world settings often involve repeated serial negotiations between the same parties. Repeated negotiations are interesting because they allow the discovery of mutually beneficial solutions that don’t exist within the confines of a single negotiation. This paper introduces the notion of Pareto efficiency over time to formalize this notion of value-creation through repeated interactions. We review literature from human negotiation research and identify a dialog strategy, favors and ledgers, that facilitates this process. As part of a longer-term effort to build intelligent virtual humans that can train human negotiators, we create a conversational agent that instantiates this strategy, and assess its effectiveness with human users, using the established Colored Trails negotiation testbed. In an empirical study involving a series of repeated negotiations, we show that humans are more likely to discover Pareto optimal solutions overtime when matched with our favor-seeking agent. Further, an agent that asks for favors during early negotiations, regardless of whether these favors are ever repaid, leads participants to discover more joint value in later negotiations, even under the traditional definition of Pareto optimality within a single negotiation. Further, agents that match their words with deeds (repay their favors) create the most value for themselves. We discuss the implications of these findings for agents that engage in long-term interactions with human users.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Automated negotiation research focuses on getting the most value from a single negotiation, yet real-world settings often involve repeated serial negotiations between the same parties. Repeated negotiations are interesting because they allow the discovery of mutually beneficial solutions that don’t exist within the confines of a single negotiation. This paper introduces the notion of Pareto efficiency over time to formalize this notion of value-creation through repeated interactions. We review literature from human negotiation research and identify a dialog strategy, favors and ledgers, that facilitates this process. As part of a longer-term effort to build intelligent virtual humans that can train human negotiators, we create a conversational agent that instantiates this strategy, and assess its effectiveness with human users, using the established Colored Trails negotiation testbed. In an empirical study involving a series of repeated negotiations, we show that humans are more likely to discover Pareto optimal solutions overtime when matched with our favor-seeking agent. Further, an agent that asks for favors during early negotiations, regardless of whether these favors are ever repaid, leads participants to discover more joint value in later negotiations, even under the traditional definition of Pareto optimality within a single negotiation. Further, agents that match their words with deeds (repay their favors) create the most value for themselves. We discuss the implications of these findings for agents that engage in long-term interactions with human users.
506.
Saito, Shunsuke; Huang, Zeng; Natsume, Ryota; Morishima, Shigeo; Kanazawa, Angjoo; Li, Hao
PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization Journal Article
In: arXiv:1905.05172 [cs], 2015.
@article{saito_pifu_2015,
title = {PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization},
author = {Shunsuke Saito and Zeng Huang and Ryota Natsume and Shigeo Morishima and Angjoo Kanazawa and Hao Li},
url = {http://arxiv.org/abs/1905.05172},
year = {2015},
date = {2015-05-01},
journal = {arXiv:1905.05172 [cs]},
abstract = {We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pixels of 2D images with the global context of their corresponding 3D object. Using PIFu, we propose an end-to-end deep learning method for digitizing highly detailed clothed humans that can infer both 3D surface and texture from a single image, and optionally, multiple input images. Highly intricate shapes, such as hairstyles, clothing, as well as their variations and deformations can be digitized in a unified way. Compared to existing representations used for 3D deep learning, PIFu can produce high-resolution surfaces including largely unseen regions such as the back of a person. In particular, it is memory efficient unlike the voxel representation, can handle arbitrary topology, and the resulting surface is spatially aligned with the input image. Furthermore, while previous techniques are designed to process either a single image or multiple views, PIFu extends naturally to arbitrary number of views. We demonstrate high-resolution and robust reconstructions on real world images from the DeepFashion dataset, which contains a variety of challenging clothing types. Our method achieves state-of-the-art performance on a public benchmark and outperforms the prior work for clothed human digitization from a single image.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pixels of 2D images with the global context of their corresponding 3D object. Using PIFu, we propose an end-to-end deep learning method for digitizing highly detailed clothed humans that can infer both 3D surface and texture from a single image, and optionally, multiple input images. Highly intricate shapes, such as hairstyles, clothing, as well as their variations and deformations can be digitized in a unified way. Compared to existing representations used for 3D deep learning, PIFu can produce high-resolution surfaces including largely unseen regions such as the back of a person. In particular, it is memory efficient unlike the voxel representation, can handle arbitrary topology, and the resulting surface is spatially aligned with the input image. Furthermore, while previous techniques are designed to process either a single image or multiple views, PIFu extends naturally to arbitrary number of views. We demonstrate high-resolution and robust reconstructions on real world images from the DeepFashion dataset, which contains a variety of challenging clothing types. Our method achieves state-of-the-art performance on a public benchmark and outperforms the prior work for clothed human digitization from a single image.
507.
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.
508.
Scherer, Stefan; Morency, Louis-Philippe; Gratch, Jonathan; Pestian, John
REDUCED VOWEL SPACE IS A ROBUST INDICATOR OF PSYCHOLOGICAL DISTRESS: A CROSS-CORPUS ANALYSIS Proceedings Article
In: Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 4789–4793, IEEE, Brisbane, Australia, 2015.
@inproceedings{scherer_reduced_2015,
title = {REDUCED VOWEL SPACE IS A ROBUST INDICATOR OF PSYCHOLOGICAL DISTRESS: A CROSS-CORPUS ANALYSIS},
author = {Stefan Scherer and Louis-Philippe Morency and Jonathan Gratch and John Pestian},
url = {http://ict.usc.edu/pubs/REDUCED%20VOWEL%20SPACE%20IS%20A%20ROBUST%20INDICATOR%20OF%20PSYCHOLOGICAL%20DISTRESS-A%20CROSS-CORPUS%20ANALYSIS.pdf},
year = {2015},
date = {2015-04-01},
booktitle = {Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)},
pages = {4789–4793},
publisher = {IEEE},
address = {Brisbane, Australia},
abstract = {Reduced frequency range in vowel production is a well documented speech characteristic of individuals’ with psychological and neurological disorders. Depression is known to influence motor control and in particular speech production. The assessment and documentation of reduced vowel space and associated perceived hypoarticulation and reduced expressivity often rely on subjective assessments. Within this work, we investigate an automatic unsupervised machine learning approach to assess a speaker’s vowel space within three distinct speech corpora and compare observed vowel space measures of subjects with and without psychological conditions associated with psychological distress, namely depression, post-traumatic stress disorder (PTSD), and suicidality. Our experiments are based on recordings of over 300 individuals. The experiments show a significantly reduced vowel space in conversational speech for depression, PTSD, and suicidality. We further observe a similar trend of reduced vowel space for read speech. A possible explanation for a reduced vowel space is psychomotor retardation, a common symptom of depression that influences motor control and speech production.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Reduced frequency range in vowel production is a well documented speech characteristic of individuals’ with psychological and neurological disorders. Depression is known to influence motor control and in particular speech production. The assessment and documentation of reduced vowel space and associated perceived hypoarticulation and reduced expressivity often rely on subjective assessments. Within this work, we investigate an automatic unsupervised machine learning approach to assess a speaker’s vowel space within three distinct speech corpora and compare observed vowel space measures of subjects with and without psychological conditions associated with psychological distress, namely depression, post-traumatic stress disorder (PTSD), and suicidality. Our experiments are based on recordings of over 300 individuals. The experiments show a significantly reduced vowel space in conversational speech for depression, PTSD, and suicidality. We further observe a similar trend of reduced vowel space for read speech. A possible explanation for a reduced vowel space is psychomotor retardation, a common symptom of depression that influences motor control and speech production.
509.
Fyffe, Graham; Debevec, Paul
Single-Shot Reflectance Measurement from Polarized Color Gradient Illumination Proceedings Article
In: Preceedings of ICCP 2015, pp. 1–10, IEEE, Houston, Texas, 2015.
@inproceedings{fyffe_single-shot_2015,
title = {Single-Shot Reflectance Measurement from Polarized Color Gradient Illumination},
author = {Graham Fyffe and Paul Debevec},
url = {http://ict.usc.edu/pubs/Single-Shot%20Reflectance%20Measurement%20from%20Polarized%20Color%20Gradient%20Illumination.pdf},
year = {2015},
date = {2015-04-01},
booktitle = {Preceedings of ICCP 2015},
pages = {1–10},
publisher = {IEEE},
address = {Houston, Texas},
abstract = {We present a method for acquiring the per-pixel diffuse albedo, specular albedo, and surface normal maps of a subject at a single instant in time. The method is single shot, requiring no optical flow, and per-pixel, making no assumptions regarding albedo statistics or surface connectivity. We photograph the subject inside a spherical illumination device emitting a static lighting pattern of vertically polarized RGB color gradients aligned with the XYZ axes, and horizontally polarized RGB color gradients in versely aligned with the XYZ axes. We capture simultaneous photographs using one of two possible setups: a single view setup using a coaxially aligned camera pair with a polarizing beam splitter, and a multi-view stereo setup with different orientations of linear polarizing filters placed on the cameras, enabling high-quality geometry reconstruction. From this lighting we derive full-color diffuse albedo, single-channel specular albedo suitable for dielectric materials, and polarization-preserving surface normals which are free of corruption from subsurface scattering. We provide simple formulae to estimate the diffuse albedo, specular albedo, and surface normal maps in the single-view and multi-view cases and show error bounds which are small for many common subjects including faces.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a method for acquiring the per-pixel diffuse albedo, specular albedo, and surface normal maps of a subject at a single instant in time. The method is single shot, requiring no optical flow, and per-pixel, making no assumptions regarding albedo statistics or surface connectivity. We photograph the subject inside a spherical illumination device emitting a static lighting pattern of vertically polarized RGB color gradients aligned with the XYZ axes, and horizontally polarized RGB color gradients in versely aligned with the XYZ axes. We capture simultaneous photographs using one of two possible setups: a single view setup using a coaxially aligned camera pair with a polarizing beam splitter, and a multi-view stereo setup with different orientations of linear polarizing filters placed on the cameras, enabling high-quality geometry reconstruction. From this lighting we derive full-color diffuse albedo, single-channel specular albedo suitable for dielectric materials, and polarization-preserving surface normals which are free of corruption from subsurface scattering. We provide simple formulae to estimate the diffuse albedo, specular albedo, and surface normal maps in the single-view and multi-view cases and show error bounds which are small for many common subjects including faces.
510.
Shim, Han Suk; Park, Sunghyun; Chatterjee, Moitreya; Scherer, Stefan; Sagae, Kenji; Morency, Louis-Philippe
ACOUSTIC AND PARA-VERBAL INDICATORS OF PERSUASIVENESS IN SOCIAL MULTIMEDIA Proceedings Article
In: Proceeding of ICASSP 2015, pp. 2239 – 2243, IEEE, Brisbane, Australia, 2015.
@inproceedings{shim_acoustic_2015,
title = {ACOUSTIC AND PARA-VERBAL INDICATORS OF PERSUASIVENESS IN SOCIAL MULTIMEDIA},
author = {Han Suk Shim and Sunghyun Park and Moitreya Chatterjee and Stefan Scherer and Kenji Sagae and Louis-Philippe Morency},
url = {http://ict.usc.edu/pubs/ACOUSTIC%20AND%20PARA-VERBAL%20INDICATORS%20OF%20PERSUASIVENESS%20IN%20SOCIAL%20MULTIMEDIA.pdf},
year = {2015},
date = {2015-04-01},
booktitle = {Proceeding of ICASSP 2015},
pages = {2239 – 2243},
publisher = {IEEE},
address = {Brisbane, Australia},
abstract = {Persuasive communication and interaction play an important and pervasive role in many aspects of our lives. With the rapid growth of social multimedia websites such as YouTube, it has become more important and useful to understand persuasiveness in the context of online social multimedia content. In this paper, we present our resultsof conducting various analyses of persuasiveness in speech with our multimedia corpus of 1,000 movie review videos obtained from ExpoTV.com, a popular social multimedia website. Our experiments firstly show that a speaker’s level of persuasiveness can be predicted from acoustic characteristics and para-verbal cues related to speech fluency. Secondly, we show that taking acoustic cues in different time periods of a movie review can improve the performance of predicting a speaker’s level of persuasiveness. Lastly, we show that a speaker’s positive or negative attitude toward a topic influences the prediction performance as well.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Persuasive communication and interaction play an important and pervasive role in many aspects of our lives. With the rapid growth of social multimedia websites such as YouTube, it has become more important and useful to understand persuasiveness in the context of online social multimedia content. In this paper, we present our resultsof conducting various analyses of persuasiveness in speech with our multimedia corpus of 1,000 movie review videos obtained from ExpoTV.com, a popular social multimedia website. Our experiments firstly show that a speaker’s level of persuasiveness can be predicted from acoustic characteristics and para-verbal cues related to speech fluency. Secondly, we show that taking acoustic cues in different time periods of a movie review can improve the performance of predicting a speaker’s level of persuasiveness. Lastly, we show that a speaker’s positive or negative attitude toward a topic influences the prediction performance as well.
511.
Wang, Ning; Pynadath, David V.; Marsella, Stacy C.
Subjective Perceptions in Wartime Negotiation Journal Article
In: IEEE Transactions on Affective Computing, vol. 6, no. 2, pp. 118–126, 2015, ISSN: 1949-3045.
@article{wang_subjective_2015,
title = {Subjective Perceptions in Wartime Negotiation},
author = {Ning Wang and David V. Pynadath and Stacy C. Marsella},
url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=6975149},
doi = {10.1109/TAFFC.2014.2378312},
issn = {1949-3045},
year = {2015},
date = {2015-04-01},
journal = {IEEE Transactions on Affective Computing},
volume = {6},
number = {2},
pages = {118–126},
abstract = {The prevalence of negotiation in social interaction has motivated researchers to develop virtual agents that can understand, facilitate, teach and even carry out negotiations. While much of this research has analyzed how to maximize the objective outcome, there is a growing body of work demonstrating that subjective perceptions of the outcome also play a critical role in human negotiation behavior. People derive subjective value from not only the outcome, but also from the process by which they achieve that outcome, from their relationship with their negotiation partner, etc. The affective responses evoked by these subjective valuations can be very different from what would be evoked by the objective outcome alone. We investigate such subjective valuations within human-agent negotiation in four variations of a wartime negotiation game. We observe that the objective outcomes of these negotiations are not strongly correlated with the human negotiators’ subjective perceptions, as measured by the Subjective Value Index. We examine the game dynamics and agent behaviors to identify features that induce different subjective values in the participants. We thus are able to identify characteristics of the negotiation process and the agents’ behavior that most impact people’s subjective valuations in our wartime negotiation games.⬚},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The prevalence of negotiation in social interaction has motivated researchers to develop virtual agents that can understand, facilitate, teach and even carry out negotiations. While much of this research has analyzed how to maximize the objective outcome, there is a growing body of work demonstrating that subjective perceptions of the outcome also play a critical role in human negotiation behavior. People derive subjective value from not only the outcome, but also from the process by which they achieve that outcome, from their relationship with their negotiation partner, etc. The affective responses evoked by these subjective valuations can be very different from what would be evoked by the objective outcome alone. We investigate such subjective valuations within human-agent negotiation in four variations of a wartime negotiation game. We observe that the objective outcomes of these negotiations are not strongly correlated with the human negotiators’ subjective perceptions, as measured by the Subjective Value Index. We examine the game dynamics and agent behaviors to identify features that induce different subjective values in the participants. We thus are able to identify characteristics of the negotiation process and the agents’ behavior that most impact people’s subjective valuations in our wartime negotiation games.⬚
512.
Andreatta, Pamela; Klotz, Jessica J.; Madsen, James M.; Hurst, Charles G.; Talbot, Thomas B.
Outcomes From Two Forms of Training for First-Responder Competency in Cholinergic Crisis Management Journal Article
In: Military Medicine, vol. 180, no. 4, pp. 468–474, 2015, ISSN: 0026-4075, 1930-613X.
@article{andreatta_outcomes_2015,
title = {Outcomes From Two Forms of Training for First-Responder Competency in Cholinergic Crisis Management},
author = {Pamela Andreatta and Jessica J. Klotz and James M. Madsen and Charles G. Hurst and Thomas B. Talbot},
url = {http://ict.usc.edu/pubs/Outcomes%20From%20Two%20Forms%20of%20Training%20for%20First-Responder%20Competency%20in%20Cholinergic%20Crisis%20Management.pdf},
doi = {10.7205/MILMED-D-14-00290},
issn = {0026-4075, 1930-613X},
year = {2015},
date = {2015-04-01},
journal = {Military Medicine},
volume = {180},
number = {4},
pages = {468–474},
abstract = {Military and civilian first responders must be able to recognize and effectively manage mass disaster casualties. Clinical management of injuries resulting from nerve agents provides different challenges for first responders than those of conventional weapons. We evaluated the impact of a mixed-methods training program on competency acquisition in cholinergic crisis clinical management using multimedia with either live animal or patient actor examples, and hands-on practice using SimMan3G mannequin simulators. A purposively selected sample of 204 civilian and military first responders who had not previously completed nerve agent training were assessed pre- and post-training for knowledge, performance, self-efficacy, and affective state. We conducted analysis of variance with repeated measures; statistical significance p textbackslashtextbackslashtextless 0.05. Both groups had significant performance improvement across all assessment dimensions: knowledge textbackslashtextbackslashtextgreater 20%, performance textbackslashtextbackslashtextgreater 50%, self-efficacy textbackslashtextbackslashtextgreater 34%, and affective state textbackslashtextbackslashtextgreater 15%. There were no significant differences between the live animal and patient actor groups. These findings could aid in the specification of training for first-responder personnel in military and civilian service. Although less comprehensive than U.S. Army Medical Research Institute of Chemical Defense courses, the training outcomes associated with this easily distributed program demonstrate its value in increasing the competency of first responders in recognizing and managing a mass casualty cholinergic event.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Military and civilian first responders must be able to recognize and effectively manage mass disaster casualties. Clinical management of injuries resulting from nerve agents provides different challenges for first responders than those of conventional weapons. We evaluated the impact of a mixed-methods training program on competency acquisition in cholinergic crisis clinical management using multimedia with either live animal or patient actor examples, and hands-on practice using SimMan3G mannequin simulators. A purposively selected sample of 204 civilian and military first responders who had not previously completed nerve agent training were assessed pre- and post-training for knowledge, performance, self-efficacy, and affective state. We conducted analysis of variance with repeated measures; statistical significance p textbackslashtextbackslashtextless 0.05. Both groups had significant performance improvement across all assessment dimensions: knowledge textbackslashtextbackslashtextgreater 20%, performance textbackslashtextbackslashtextgreater 50%, self-efficacy textbackslashtextbackslashtextgreater 34%, and affective state textbackslashtextbackslashtextgreater 15%. There were no significant differences between the live animal and patient actor groups. These findings could aid in the specification of training for first-responder personnel in military and civilian service. Although less comprehensive than U.S. Army Medical Research Institute of Chemical Defense courses, the training outcomes associated with this easily distributed program demonstrate its value in increasing the competency of first responders in recognizing and managing a mass casualty cholinergic event.
513.
Jones, Andrew; Unger, Jonas; Nagano, Koki; Busch, Jay; Yu, Xueming; Peng, Hsuan-Yueh; Alexander, Oleg; Debevec, Paul
Building a Life-Size Automultiscopic Display Using Consumer Hardware Proceedings Article
In: Proceedings of GPU Technology Conference, San Jose, CA, 2015.
@inproceedings{jones_building_2015,
title = {Building a Life-Size Automultiscopic Display Using Consumer Hardware},
author = {Andrew Jones and Jonas Unger and Koki Nagano and Jay Busch and Xueming Yu and Hsuan-Yueh Peng and Oleg Alexander and Paul Debevec},
url = {http://ict.usc.edu/pubs/Building%20a%20Life-Size%20Automultiscopic%20Display%20Using%20Consumer%20Hardware.pdf},
year = {2015},
date = {2015-03-01},
booktitle = {Proceedings of GPU Technology Conference},
address = {San Jose, CA},
abstract = {Automultiscopic displays allow multiple users to experience 3D content without the hassle of special glasses or head gear. Such displays generate many simultaneous images with high-angular density, so that each eye perceives a distinct and different view. This presents a unique challenge for content acquisition and rendering. In this talk, we explain how to build an automultiscopic display using off-the-shelf projectors, video-splitters, and graphics cards. We also present a GPU-based algorithm for rendering a large numbers of views from a sparse array of video cameras.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Automultiscopic displays allow multiple users to experience 3D content without the hassle of special glasses or head gear. Such displays generate many simultaneous images with high-angular density, so that each eye perceives a distinct and different view. This presents a unique challenge for content acquisition and rendering. In this talk, we explain how to build an automultiscopic display using off-the-shelf projectors, video-splitters, and graphics cards. We also present a GPU-based algorithm for rendering a large numbers of views from a sparse array of video cameras.
514.
DeVault, David; Mell, Jonathan; Gratch, Jonathan
Toward Natural Turn-Taking in a Virtual Human Negotiation Agent Proceedings Article
In: AAAI Spring Symposium on Turn-taking and Coordination in Human-Machine Interaction, pp. 2–9, AAAI Press, Palo Alto, California, 2015.
@inproceedings{devault_toward_2015,
title = {Toward Natural Turn-Taking in a Virtual Human Negotiation Agent},
author = {David DeVault and Jonathan Mell and Jonathan Gratch},
url = {http://ict.usc.edu/pubs/Toward%20Natural%20Turn-Taking%20in%20a%20Virtual%20Human%20Negotiation%20Agent.pdf},
year = {2015},
date = {2015-03-01},
booktitle = {AAAI Spring Symposium on Turn-taking and Coordination in Human-Machine Interaction},
pages = {2–9},
publisher = {AAAI Press},
address = {Palo Alto, California},
abstract = {In this paper we assess our progress toward creating a virtual human negotiation agent with fluid turn-taking skills. To facilitate the design of this agent, we have collected a corpus of human-human negotiation roleplays as well as a corpus of Wizard-controlled human-agent negotiations in the same roleplay scenario.We compare the natural turn-taking behavior in our human-human corpus with that achieved in our Wizard-of-Oz corpus, and quantify our virtual human’s turn-taking skills using a combination of subjective and objective metrics. We also discuss our design for a Wizard user interface to support real-time control of the virtual human’s turntaking and dialogue behavior, and analyze our wizard’s usage of this interface.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper we assess our progress toward creating a virtual human negotiation agent with fluid turn-taking skills. To facilitate the design of this agent, we have collected a corpus of human-human negotiation roleplays as well as a corpus of Wizard-controlled human-agent negotiations in the same roleplay scenario.We compare the natural turn-taking behavior in our human-human corpus with that achieved in our Wizard-of-Oz corpus, and quantify our virtual human’s turn-taking skills using a combination of subjective and objective metrics. We also discuss our design for a Wizard user interface to support real-time control of the virtual human’s turntaking and dialogue behavior, and analyze our wizard’s usage of this interface.
515.
Ward, Nigel G.; DeVault, David
Ten Challenges in Highly-Interactive Dialog Systems Proceedings Article
In: Proceedings of AAAI 2015 Spring Symposium, Palo Alto, CA, 2015.
@inproceedings{ward_ten_2015,
title = {Ten Challenges in Highly-Interactive Dialog Systems},
author = {Nigel G. Ward and David DeVault},
url = {http://ict.usc.edu/pubs/Ten%20Challenges%20in%20Highly-Interactive%20Dialog%20Systems.pdf},
year = {2015},
date = {2015-03-01},
booktitle = {Proceedings of AAAI 2015 Spring Symposium},
address = {Palo Alto, CA},
abstract = {Systems capable of highly-interactive dialog have recently been developed in several domains. This paper considers how to build on these successes to make systems more robust, easier to develop, more adaptable, and more scientifically significant.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Systems capable of highly-interactive dialog have recently been developed in several domains. This paper considers how to build on these successes to make systems more robust, easier to develop, more adaptable, and more scientifically significant.
516.
Casas, Dan; Alexander, Oleg; Feng, Andrew W.; Fyffe, Graham; Ichikari, Ryosuke; Debevec, Paul; Wang, Rhuizhe; Suma, Evan; Shapiro, Ari
Rapid Photorealistic Blendshapes from Commodity RGB-D Sensors Proceedings Article
In: Proceedings of the 19th Symposium on Interactive 3D Graphics and Games, pp. 134–134, ACM Press, San Francisco, CA, 2015, ISBN: 978-1-4503-3392-4.
@inproceedings{casas_rapid_2015,
title = {Rapid Photorealistic Blendshapes from Commodity RGB-D Sensors},
author = {Dan Casas and Oleg Alexander and Andrew W. Feng and Graham Fyffe and Ryosuke Ichikari and Paul Debevec and Rhuizhe Wang and Evan Suma and Ari Shapiro},
url = {http://dl.acm.org/citation.cfm?doid=2699276.2721398},
doi = {10.1145/2699276.2721398},
isbn = {978-1-4503-3392-4},
year = {2015},
date = {2015-02-01},
booktitle = {Proceedings of the 19th Symposium on Interactive 3D Graphics and Games},
pages = {134–134},
publisher = {ACM Press},
address = {San Francisco, CA},
abstract = {Creating and animating a realistic 3D human face has been an important task in computer graphics. The capability of capturing the 3D face of a human subject and reanimate it quickly will find many applications in games, training simulations, and interactive 3D graphics. In this paper, we propose a system to capture photorealistic 3D faces and generate the blendshape models automatically using only a single commodity RGB-D sensor. Our method can rapidly generate a set of expressive facial poses from a single Microsoft Kinect and requires no artistic expertise on the part of the capture subject. The system takes only a matter of seconds to capture and produce a 3D facial pose and only requires 4 minutes of processing time to transform it into a blendshape model. Our main contributions include an end-to-end pipeline for capturing and generating face blendshape models automatically, and a registration method that solves dense correspondences between two face scans by utilizing facial landmark detection and optical flow. We demonstrate the effectiveness of the proposed method by capturing 3D facial models of different human subjects and puppeteering their models in an animation system with real-time facial performance retargeting.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Creating and animating a realistic 3D human face has been an important task in computer graphics. The capability of capturing the 3D face of a human subject and reanimate it quickly will find many applications in games, training simulations, and interactive 3D graphics. In this paper, we propose a system to capture photorealistic 3D faces and generate the blendshape models automatically using only a single commodity RGB-D sensor. Our method can rapidly generate a set of expressive facial poses from a single Microsoft Kinect and requires no artistic expertise on the part of the capture subject. The system takes only a matter of seconds to capture and produce a 3D facial pose and only requires 4 minutes of processing time to transform it into a blendshape model. Our main contributions include an end-to-end pipeline for capturing and generating face blendshape models automatically, and a registration method that solves dense correspondences between two face scans by utilizing facial landmark detection and optical flow. We demonstrate the effectiveness of the proposed method by capturing 3D facial models of different human subjects and puppeteering their models in an animation system with real-time facial performance retargeting.
517.
Park, Sunghyun; Scherer, Stefan; Gratch, Jonathan; Carnevale, Peter; Morency, Louis-Philippe
I Can Already Guess Your Answer: Predicting Respondent Reactions During Dyadic Negotiation Journal Article
In: IEEE Transactions on Affective Computing, vol. 6, no. 2, pp. 86 –96, 2015, ISSN: 1949-3045.
@article{park_i_2015,
title = {I Can Already Guess Your Answer: Predicting Respondent Reactions During Dyadic Negotiation},
author = {Sunghyun Park and Stefan Scherer and Jonathan Gratch and Peter Carnevale and Louis-Philippe Morency},
url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7024926},
doi = {10.1109/TAFFC.2015.2396079},
issn = {1949-3045},
year = {2015},
date = {2015-01-01},
journal = {IEEE Transactions on Affective Computing},
volume = {6},
number = {2},
pages = {86 –96},
abstract = {Negotiation is a component deeply ingrained in our daily lives, and it can be challenging for a person to predict the respondent’s reaction (acceptance or rejection) to a negotiation offer. In this work, we focus on finding acoustic and visual behavioral cues that are predictive of the respondent’s immediate reactions using a face-to-face negotiation dataset, which consists of 42 dyadic interactions in a simulated negotiation setting. We show our results of exploring 4 different sources of information, namely nonverbal behavior of the proposer, that of the respondent, mutual behavior between the interactants related to behavioral symmetry and asymmetry, and past negotiation history between the interactants. Firstly, we show that considering other sources of information (other than the nonverbal behavior of the respondent) can also have comparable performance in predicting respondent reactions. Secondly, we show that automatically extracted mutual behavioral cues of symmetry and asymmetry are predictive partially due to their capturing information of the nature of the interaction itself, whether it is cooperative or competitive. Lastly, we identify audio-visual behavioral cues that are most predictive of the respondent’s immediate reactions.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Negotiation is a component deeply ingrained in our daily lives, and it can be challenging for a person to predict the respondent’s reaction (acceptance or rejection) to a negotiation offer. In this work, we focus on finding acoustic and visual behavioral cues that are predictive of the respondent’s immediate reactions using a face-to-face negotiation dataset, which consists of 42 dyadic interactions in a simulated negotiation setting. We show our results of exploring 4 different sources of information, namely nonverbal behavior of the proposer, that of the respondent, mutual behavior between the interactants related to behavioral symmetry and asymmetry, and past negotiation history between the interactants. Firstly, we show that considering other sources of information (other than the nonverbal behavior of the respondent) can also have comparable performance in predicting respondent reactions. Secondly, we show that automatically extracted mutual behavioral cues of symmetry and asymmetry are predictive partially due to their capturing information of the nature of the interaction itself, whether it is cooperative or competitive. Lastly, we identify audio-visual behavioral cues that are most predictive of the respondent’s immediate reactions.
518.
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.
519.
Morency, Louis-Philippe; Stratou, Giota; DeVault, David; Hartholt, Arno; Lhommet, Margaux; Lucas, Gale; Morbini, Fabrizio; Georgila, Kallirroi; Scherer, Stefan; Gratch, Jonathan; Marsella, Stacy; Traum, David; Rizzo, Albert "Skip"
SimSensei Demonstration: A Perceptive Virtual Human Interviewer for Healthcare Applications Proceedings Article
In: Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI), Austin, Texas, 2015.
@inproceedings{morency_simsensei_2015,
title = {SimSensei Demonstration: A Perceptive Virtual Human Interviewer for Healthcare Applications},
author = {Louis-Philippe Morency and Giota Stratou and David DeVault and Arno Hartholt and Margaux Lhommet and Gale Lucas and Fabrizio Morbini and Kallirroi Georgila and Stefan Scherer and Jonathan Gratch and Stacy Marsella and David Traum and Albert "Skip" Rizzo},
url = {http://ict.usc.edu/pubs/SimSensei%20Demonstration%20A%20Perceptive%20Virtual%20Human%20Interviewer%20for%20Healthcare%20Applications.pdf},
year = {2015},
date = {2015-01-01},
booktitle = {Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI)},
address = {Austin, Texas},
abstract = {We present the SimSensei system, a fully automatic virtual agent that conducts interviews to assess indicators of psychological distress. We emphasize on the perception part of the system, a multimodal framework which captures and analyzes user state for both behavioral understanding and interactional purposes.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present the SimSensei system, a fully automatic virtual agent that conducts interviews to assess indicators of psychological distress. We emphasize on the perception part of the system, a multimodal framework which captures and analyzes user state for both behavioral understanding and interactional purposes.
520.
Corbin, Carina; Morbini, Fabrizio; Traum, David
Creating a Virtual Neighbor Proceedings Article
In: Proceedings of International Workshop on Spoken Dialogue Systems, Busan, South Korea, 2015.
@inproceedings{corbin_creating_2015,
title = {Creating a Virtual Neighbor},
author = {Carina Corbin and Fabrizio Morbini and David Traum},
url = {http://ict.usc.edu/pubs/Creating%20a%20Virtual%20Neighbor.pdf},
year = {2015},
date = {2015-01-01},
booktitle = {Proceedings of International Workshop on Spoken Dialogue Systems},
address = {Busan, South Korea},
abstract = {We present the first version of our Virtual Neighbor, who can talk with users about people employed in the same institution. The Virtual Neighbor can discuss information about employees in a medium sized company or institute with users. The system acquires information from three sources: a personnel directory database, public web pages, and through dialogue interaction. Users can interact through face to face spoken dialogue, using components from the ICT Virtual human toolkit, or via a chat interface.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present the first version of our Virtual Neighbor, who can talk with users about people employed in the same institution. The Virtual Neighbor can discuss information about employees in a medium sized company or institute with users. The system acquires information from three sources: a personnel directory database, public web pages, and through dialogue interaction. Users can interact through face to face spoken dialogue, using components from the ICT Virtual human toolkit, or via a chat interface.
Filter
Sorry, no publications matched your criteria.