Select a calendar:
Filter November Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for November 14, 2017
-
PhD Defense- Dehua Cheng
Tue, Nov 14, 2017 @ 10:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
University Calendar
Tuesday, November, 14th, 10 a.m. to 12 p.m., PHE 223
Title: Improving machine learning algorithms with efficient data relevance discovery
Abstract:
This is the era of big data, where both challenges and opportunities lie ahead for the machine learning research. The data are created nowadays at an unprecedented pace with an unignorable cost in collecting, storing, and computing with the current scale of data. As the computational power that we possess gradually plateaus, it is an ever-increasing challenge to fully utilize the wealth of big data, where better data reduction techniques and scalable algorithms are the keys to a solution. We observe that to answer a certain query, the data are not equally important. Based on the models and the query, we provide efficient access to the numerical scores of the data points that represent their relevance in the current task. It enables us to wisely devote the computation resources to the important data, which improves the scalability and the reliability. We present our work under three applications: 1) tensor CP decomposition, 2) random-walk matrix-polynomial sparsification, where we provide an efficient access to the statistical leverage score for a faster numerical routine; and 3) matrix completability analysis, where we analyze the underlying completability structure for a more reliable estimation.
Location: Charles Lee Powell Hall (PHE) - 223
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
MASCLE Machine Learning Seminar: Robert Schapire (Microsoft Research NYC) - The Contextual Bandits Problem: Techniques for Learning to Make High-Reward Decisions
Tue, Nov 14, 2017 @ 03:30 PM - 04:50 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Robert Schapire, Microsoft Research NYC
Talk Title: The Contextual Bandits Problem: Techniques for Learning to Make High-Reward Decisions
Series: NVIDIA Distinguished Lecture Series in Machine Learning
Abstract: This lecture satisfies requirements for CSCI 591: Research Colloquium.
We consider how to learn through experience to make intelligent decisions. In the generic setting, called the contextual bandits problem, the learner must repeatedly decide which action to take in response to an observed context, and is then permitted to observe the received reward, but only for the chosen action. The goal is to learn to behave nearly as well as the best policy (or decision rule) in some possibly very large and rich space of candidate policies. This talk will describe progress on developing general methods for this problem and some of its variants.
Biography: Robert Schapire is a Principal Researcher at Microsoft Research in New York City. He received his PhD from MIT in 1991. After a short post-doc at Harvard, he joined the technical staff at AT&T Labs (formerly AT&T Bell Laboratories) in 1991. In 2002, he became a Professor of Computer Science at Princeton University. He joined Microsoft Research in 2014. His awards include the 1991 ACM Doctoral Dissertation Award, the 2003 Gödel Prize, and the 2004 Kanelakkis Theory and Practice Award (both of the last two with Yoav Freund). He is a fellow of the AAAI, and a member of both the National Academy of Engineering and the National Academy of Sciences. His main research interest is in theoretical and applied machine learning, with particular focus on boosting, online learning, game theory, and maximum entropy.
Host: Haipeng Luo
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Computer Science Department
-
CS Colloquium: Dr. Brian Milch (Google) - Combining Probabilistic and Neural Approaches for Text Classification
Tue, Nov 14, 2017 @ 05:00 PM - 06:20 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr. Brian Milch, Google
Talk Title: Combining Probabilistic and Neural Approaches for Text Classification
Series: CS Colloquium
Abstract: This lecture satisfies requirements for CSCI 591: Research Colloquium.
In the Semantic Signals group at Google Los Angeles, we build classifiers that label text with hundreds of human-defined categories across dozens of languages. Labeled training data is sparse, so we've found it essential to incorporate unsupervised learning methods that take advantage of unlabeled text. One of our tools is a probabilistic topic model that learns discrete "clusters" to explain word co-occurrence patterns in a large corpus, and then identifies the clusters that best explain a new document. Another tool is a neural net that learns embeddings of individual words in a continuous space. I'll discuss how these approaches play complementary roles in our text classification pipeline.
Biography: Brian Milch is a software engineer at Google's Los Angeles office. He received a B.S. in Symbolic Systems from Stanford University in 2000, and a Ph.D. in Computer Science from U.C. Berkeley in 2006. He then spent two years as a post-doctoral researcher at MIT before joining Google in 2008. He has contributed to Google production systems for spelling correction, transliteration, and semantic modeling of text.
Host: Fei Sha
Location: Seeley G. Mudd Building (SGM) - 124
Audiences: Everyone Is Invited
Contact: Computer Science Department
