Conferences, Lectures, & Seminars
Events for March
-
INFORMATION DROPOUT: LEARNING OPTIMAL REPRESENTATIONS THROUGH NOISY COMPUTATION
Tue, Mar 07, 2017 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Alessandro Achille, UCLA
Talk Title: INFORMATION DROPOUT: LEARNING OPTIMAL REPRESENTATIONS THROUGH NOISY COMPUTATION
Series: Natural Language Seminar
Abstract: The cross-entropy loss commonly used in deep learning is closely related to the information theoretic properties defining an optimal representation of the data, but does not enforce some of the key properties. We show that this can be solved by adding a regularization term, which is in turn related to injecting multiplicative noise in the activations of a Deep Neural Network, a special case of which is the common practice of dropout. Our regularized loss function can be efficiently minimized using Information Dropout, a generalization of dropout rooted in information theoretic principles that automatically adapts to the data and can better exploit architectures of limited capacity.
When the task is the reconstruction of the input, we show that our loss function yields a Variational Autoencoder as a special case, thus providing a link between representation learning, information theory and variational inference. Finally, we prove that we can promote the creation of disentangled representations of the input simply by enforcing a factorized prior, a fact that has been also observed empirically in recent work.
Our experiments validate the theoretical intuitions behind our method, and we find that Information Dropout achieves a comparable or better generalization performance than binary dropout, especially on smaller models, since it can automatically adapt the noise structure to the architecture of the network, as well as to the test sample.
Biography: Alessandro Achille is a PhD student in Computer Science at UCLA, working with Prof. Stefano Soatto. He focuses on variational inference, representation learning, and their applications to deep learning and computer vision. Before coming to UCLA, he obtained a Master's degree in Pure Math at the Scuola Normale Superiore in Pisa, where he studied model theory and algebraic topology with Prof. Alessandro Berarducci.
Host: Greg Ver Steeg
More Info: https://arxiv.org/abs/1611.01353
Location: Information Science Institute (ISI) - 6th Flr -CR#689 (ISI/Marina Del Rey)
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: https://arxiv.org/abs/1611.01353
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Learning agents that interact with humans
Fri, Mar 10, 2017 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: He He, Stanford Univ.
Talk Title: Learning agents that interact with humans
Series: Natural Language Seminar
Abstract: The future of virtual assistants, self driving cars, and smart homes require intelligent agents that work intimately with users. Instead of passively following orders given by users, an interactive agent must actively collaborate with people through communication, coordination, and user adaptation. In this talk, I will present our recent work towards building agents that interact with humans. First, we propose a symmetric collaborative dialogue setting in which two agents, each with some private knowledge, must communicate in natural language to achieve a common goal. We present a human-human dialogue dataset that poses new challenges to existing models, and propose a neural model with dynamic knowledge graph embedding. Second, we study the user-adaptation problem in quizbowl - a competitive, incremental question answering game. We show that explicitly modeling of different human behavior leads to more effective policies that exploits sub optimal players. I will conclude by discussing opportunities and open questions in learning interactive agents.
Biography: He He is a post-doc at Stanford University, working with Percy Liang. Prior to Stanford, she earned her Ph.D. in Computer Science at the University of Maryland, College Park, advised by Hal Daume III and Jordan Boyd Graber. Her interests are at the interface of machine learning and natural language processing. She develops algorithms that acquire information dynamically and do inference incrementally, with an emphasis on problems in natural language processing. She has worked on dependency parsing, simultaneous machine translation, question answering, and more recently dialogue systems.
Host: Marjan Ghazvininejad and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Heterogeneous Attribute Embedding and Sequence Modeling for Recommendation with Implicit Feedback
Fri, Mar 17, 2017 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Kuan Liu, USC/ISI
Talk Title: Heterogeneous Attribute Embedding and Sequence Modeling for Recommendation with Implicit Feedback
Series: Natural Language Seminar
Abstract: Incorporating implicit feedback into a recommender system is a challenging problem due to sparse and noisy observations. I will present our approaches that exploit heterogeneous attributes and sequence properties within the observations. We build a neural network framework to embed heterogeneous attributes in an end-to-end fashion, and apply the framework to three sequence-based models. Our methods achieve significant improvements on four large scale datasets compared to state-of-the-art baseline models 30 to 90 percent relative increase in NDCG. Experimental results show that attribute embedding and sequence modeling both lead to improvements and, further, that our novel output attribute layer plays a crucial role. I will conclude with our exploratory studies that investigate why sequence modeling works well in recommendation systems and advocate its use for large scale recommendation tasks.
Biography: Kuan Liu is a fifth year Ph.D. student at ISI/USC working with Prof. Prem Natarajan. Before that, He received a bachelor degree from Tsinghua University with a major in Computer Science. His research interests include machine learning, large scale optimization, deep learning, and applications to recommender systems, network analysis.
Host: Marjan Ghazvininejad and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
AI Seminar
Mon, Mar 20, 2017 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Xiang Ren, Computer Science PhD candidate at University of Illinois at UrbanaChampaign
Talk Title: EFFORT-LIGHT STRUCTMINE: TURNING MASSIVE CORPORA INTO STRUCTURES
Series: Recruitng Seminar
Abstract: The realworld data, though massive, are hard for machines to resolve as they are largely unstructured and in the form of natural-language text. One of the grand challenges is to turn such massive corpora into machine-actionable structures. Yet, most existing systems have heavy reliance on human effort in the process of structuring various corpora, slowing down the development of downstream applications.
In this talk, I will introduce a data-driven framework, EffortLight StructMine, that extracts structured facts from massive corpora without explicit human labeling effort. In particular, I will discuss how to solve three structure mining tasks under Effort-Light StructMine framework: from identifying typed entities in text, to fine-grained entity typing, to extracting typed relationships between entities. Together, these three solutions form a clear roadmap for turning a massive corpus into a structured network to represent its factual knowledge. Finally, I will share some directions towards mining corpus-specific structured networks for knowledge discovery.
Biography: Xiang Ren is a Computer Science PhD candidate at University of Illinois at Urbana-Champaign, working with Jiawei Han and the Data and Information System DAIS Research Lab. The research Xiang develops data-driven methods for turning unstructured text data into machine-actionable structures. More broadly, his research interests span data mining, machine learning, and natural language processing, with a focus on making sense of massive text corpora. His research has been recognized with several prestigious awards including a Google PhD Fellowship, a Yahoo!-DAIS Research Excellence Award, and a C. W. Gear Outstanding Graduate Student Award from UIUC Computer Science. Technologies he developed has been transferred to US Army Research Lab, NIH, Microsoft, Yelp and TripAdvisor
Host: Craig Knoblock
Webcast: http://webcastermshd.isi.edu/Mediasite/Play/6b83d48fc61f4e398d8d8bbdff0004e01dLocation: Information Science Institute (ISI) - 11th Floor Large CR #1135
WebCast Link: http://webcastermshd.isi.edu/Mediasite/Play/6b83d48fc61f4e398d8d8bbdff0004e01d
Audiences: Everyone Is Invited
Contact: Alma Nava / Information Sciences Institute
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
AI Seminar: EVOLUTION OF NEURAL NETWORKS
Fri, Mar 24, 2017 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Risto Miikkulainen, Univ. of Texas
Talk Title: EVOLUTION OF NEURAL NETWORKS
Series: Artificial Intelligence Seminar
Abstract: Evolution of artificial neural networks has recently emerged as a powerful technique both in deep networks and reinforcementlearning. While the performance of deep learning networks depends crucially on the network architecture; with neuroevolution, it ispossible to discover such architectures automatically. While reinforcement learning works well when the environment is fully observable, neuroevolution makes it possible to disambiguate hidden state through memory. In this tutorial, I will review 1 neuroevolution methods that evolve fixed-topology networks, network topologies, and network construction processes, 2 ways of combining gradient-based training with evolutionary methods, and 3 applications of neuroevolution to control, robotics, artificial life, and games.
Biography: Risto Miikkulainen is a Professor of Computer Science at the University of Texas at Austin and a Research Fellow at Sentient Technologies, Inc. He received an M.S. in Engineering from the Helsinki University of Technology, Finland, in 1986, and a Ph.D. in Computer Science from UCLA in 1990. His current research focuses on methods and applications of neuroevolution, as well as neural network models of natural language processing, and self-organization of the visual cortex; he is an author of over 370 articles in these research areas. He is an IEEE Fellow, member of the Board of Governors of the Neural Network Society, and an action editor of Cognitive Systems Research and IEEE Transactions on Computational Intelligence and AI in Games.
Host: Mayank Kejriwal
Webcast: http://webcastermshd.isi.edu/Mediasite/Play/f0024e2d2140457586ec2ed6a78026b01dLocation: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
WebCast Link: http://webcastermshd.isi.edu/Mediasite/Play/f0024e2d2140457586ec2ed6a78026b01d
Audiences: Everyone Is Invited
Contact: Peter Zamar
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar - Intuitive Interactions with Black-box Machine Learning
Fri, Mar 24, 2017 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Sameer Singh, UCI
Talk Title: Intuitive Interactions with Black-box Machine Learning
Series: Natural Language Seminar
Abstract: Machine learning is at the forefront of many recent advances in natural language processing, enabled in part by the sophisticated models and algorithms that have been recently introduced. However, as a consequence of this complexity, machine learning essentially acts as a black-box as far as users are concerned. It is incredibly difficult to understand, predict, or "fix" the behavior of NLP models that have been deployed. In this talk, I propose interpretable representations that allow users and machine learning models to interact with each other: enabling machine learning models to provided explanations as to why a specific prediction was made and enabling users to inject domain knowledge into machine learning. The first part of the talk introduces an approach to estimate local, interpretable explanations for black-box classifiers and describes an approach to summarize the behavior of the classifier by selecting which explanations to show to the user. I will also briefly describe work on "closing the loop", i.e. allowing users to provide feedback on the explanations to improve the model, for the task of relation extraction, an important subtask of natural language processing. In particular, we introduce approaches to both explain the relation extractor using logical statements and to inject symbolic domain knowledge into relational embeddings to improve the predictions. I present experiments to demonstrate that an interactive interface is effective in providing users an understanding of, and an ability to improve, complex black-box machine learning systems.
Biography: Sameer Singh is an Assistant Professor of Computer Science at the University of California, Irvine. He is working on large-scale and interactive machine learning applied to information extraction and natural language processing. Till recently, Sameer was a Postdoctoral Research Associate at the University of Washington. He received his PhD from the University of Massachusetts, Amherst in 2014, during which he also interned at Microsoft Research, Google Research, and Yahoo! Labs on massive-scale machine learning. He was selected as a DARPA Riser, was awarded the Adobe Research Data Science Award, won the grand prize in the Yelp dataset challenge, has been awarded the Yahoo! Key Scientific Challenges fellowship, and was a finalist for the Facebook PhD fellowship. Sameer has published more than 30 peer-reviewed papers at top-tier machine learning and natural language processing conferences and workshops.
Host: Marjan Ghazvininejad and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar - ANALYZING THE LANGUAGE OF FOOD ON SOCIAL MEDIA
Mon, Mar 27, 2017 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Stephen Kobourov , University of Arizona
Talk Title: ANALYZING THE LANGUAGE OF FOOD ON SOCIAL MEDIA
Series: Natural Language Seminar
Abstract: We investigate the predictive power behind the language of food on social media. We collect a corpus of over three million food-related posts from Twitter and demonstrate that many latent population characteristics can be directly predicted from this data: overweight rate, diabetes rate, political leaning, and home geographical location of authors. For all tasks, our language-based models significantly outperform the majority class baselines. Performance is further improved with more complex natural language processing, such as topic modeling. We analyze which textual features have most predictive power for these datasets, providing insight into the connections between the language of food, geographic locale, and community characteristics. Lastly, we design and implement an online system for real-time query and visualization of the dataset. Visualization tools, such as geo referenced heatmaps, semantics-preserving wordclouds and temporal histograms, allow us to discover more complex, global patterns mirrored in the language of food.
Biography: Stephen Kobourov is a Professor of Computer Science at the University of Arizona. He completed BS degrees in Mathematics and Computer Science at Dartmouth College in 1995, and a PhD in Computer Science at Johns Hopkins University in 2000. He has worked as a Research Scientist at AT&T Research Labs, a Hulmboldt Fellow at the University of Tubingen in Germany, and a Distinguished Fulbright Chair at Charles University in Prague.
Host: Marjan Ghazvininejad and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar Towards the Machine Comprehension of Text
Fri, Mar 31, 2017 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Danqi Chen, Stanford Univ.
Talk Title: Towards the Machine Comprehension of Text
Series: Natural Language Seminar
Abstract: In this talk, I will first present how we advance this line of research. I will show how simple models can achieve nearly state of the art performance on recent benchmarks, including the CNN Daily Mail datasets and the Stanford Question Answering Dataset. I will focus on explaining the logical structure behind these neural architectures and discussing advantage as well as limits of current approaches. Lastly I will talk about our recent work on scaling up machine comprehension systems, which attempt to answer open domain questions at the full Wikipedia scale. We demonstrate the promise of our system, as well as set up new benchmarks by evaluating on multiple existing QA datasets.
Biography: Danqi Chen is a PhD candidate in Computer Science at Stanford University, advised by Professor Christopher Manning. Her main research interests lie in deep learning for natural language processing and understanding, and she is particularly interested in the intersection between text understanding and knowledge reasoning. She has been working on machine comprehension, question answering, knowledge base population and dependency parsing. She is a recipient of Facebook fellowship and Microsoft Research Womens Fellowship and an outstanding paper award in ACL 16. Prior to Stanford, she received her BS from Tsinghua University.
Host: Marjan Ghazvininejad and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
