From macglashan at tti-c.org Tue Sep 15 14:49:23 2009 From: macglashan at tti-c.org (Julia MacGlashan) Date: Tue Sep 15 14:49:29 2009 Subject: [TTIC Colloquium] TTI-C Colloquium: Dan Roth, UIUC Message-ID: When: Monday, Sep 21 @ 2:00pm Where: 6045 S Kenwood Ave, TTI-C Conference Room #526 (5th Floor) Who: Dan Roth, Dept of Computer Science, UIUC Title: Constrained Conditional Models: Learning and Inference in Natural Language Understanding Making decisions in natural language understanding tasks often involves assigning values to sets of interdependent variables where an expressive dependency structure among these can influence, or even dictate, what assignments are possible. Structured learning problems provide one such example, but we are interested in a broader setting where multiple models are involved, global inference over these is essential, but it may not be ideal, or possible, to learn them jointly. I will present work on Constrained Conditional Models (CCMs), a framework that augments probabilistic models with declarative constraints as a way to support decisions in an expressive output space while maintaining modularity and tractability of training. The focus will be on discussing training and inference paradigms for Constrained Conditional Models, with examples drawn from natural language understanding tasks such as semantic role learning, information extraction tasks, and transliteration. Speaker Schedule: http://www.tti-c.org/ colloquium.php Contact: Karen Livescu, TTI-C klivescu@tti-c.org 834-2549 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://ttic.uchicago.edu/pipermail/colloquium/attachments/20090915/ba9126b6/attachment.htm From macglashan at tti-c.org Fri Sep 25 11:57:01 2009 From: macglashan at tti-c.org (Julia MacGlashan) Date: Fri Sep 25 11:57:12 2009 Subject: [TTIC Colloquium] New! TTI-C Seminar Series Message-ID: Research at TTI-Chicago Seminar Series Time: Tuesdays 12-1:30pm Where: TTI-C 5th Floor Conference Rooms #526-530, 6045 S Kenwood Ave TTI-Chicago is starting a weekly seminar series presenting the research done at TTI-Chicago. Every week a different TTI-C faculty member will present their research. This coming Fall Quarter Lectures will take place Tuesdays noon-1:30pm at TTI-Chicago, and the expected format is a roughly one hour lecture followed by discussion (a light lunch will also be provided). The seminar will continue in Winter Quarter at a time TBD. The lectures are intended both for students seeking research topics and advisers, and for the general TTI and UChicago communities interested in hearing what their colleagues are up to. Students, faculty and other researchers are invited to attend the seminars. Please visit http://ttic.uchicago.edu/ mailman/listinfo/ttic-seminar if you would like to be added to the seminar mailing list, to which weekly speaker announcements will be posted. Details can also be found at http://tti-c.org/tticseminar. php. The first meeting will take place Tuesday, September 29th. Prof. Nati Srebro will discuss his research on the interaction of information and computation in machine learning. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://ttic.uchicago.edu/pipermail/colloquium/attachments/20090925/48837ff6/attachment.htm From macglashan at tti-c.org Mon Sep 28 11:02:21 2009 From: macglashan at tti-c.org (Julia MacGlashan) Date: Mon Sep 28 11:02:41 2009 Subject: [TTIC Colloquium] Research at TTI-C Seminar Series Message-ID: <6FF27E67705A45EA951A70AFEE7F4FDB@jmacglDPLFYD1> STARTING TOMORROW Research at TTI-C Seminar Series Time: Tuesdays 12-1:30pm Where: TTI-C 5th Floor Conference Rooms #526-530, 6045 S Kenwood Ave TTI-C is starting a weekly seminar series presenting the research currently underway at the Institute. Every week a different TTI-C faculty member will present their research. The expected format is roughly a one hour lecture followed by discussion (a light lunch will also be provided). The seminar will continue in Winter Quarter at a time TBD. The lectures are intended both for students seeking research topics and advisers, and for the general TTI-C and UChicago communities interested in hearing what their colleagues are up to. To receive announcements about the seminar series, please subscribe to the mailing list: http://ttic.uchicago.edu/ mailman/listinfo/ttic-seminar. For details, visit http://tti-c.org/tticseminar. php or contact Nati Srebro at nati@tti-c.org. The first meeting will take place Tuesday, September 29th. Prof. Nati Srebro will discuss his research on the interaction of information and computation in machine learning. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://ttic.uchicago.edu/pipermail/colloquium/attachments/20090928/db9c8983/attachment.htm From macglashan at tti-c.org Wed Sep 30 09:42:36 2009 From: macglashan at tti-c.org (Julia MacGlashan) Date: Wed Sep 30 09:42:49 2009 Subject: [TTIC Colloquium] TTI-C Colloquium: Konstantin Makarychev, IBM Message-ID: <2B59048B52BB4526AAAD88DB4F7E662C@jmacglDPLFYD1> When: Monday, Oct 5 @ 1:00pm Where: TTI-C Conference Room #526, 6045 S Kenwood Ave, 5th Floor Who: Konstantin Makarychev (IBM T.J. Watson Research Center) Title: Sparse Rectangular Representations of Matrix Data Given a matrix of integers, how compactly can we exactly represent the matrix as a sum of weighted rectangles? More precisely, given a set of allowable rectangles, using a linear combination of how few allowed rectangles can we represent the matrix? We study two variants. Motivated by database applications, the first case allows rectangles given by the cross product of two trees. Specifically, we are given one tree whose leaves are the rows and another whose leaves are the columns. A rectangle is allowed if and only if it is the cross product between the leaf descendants of a node in the first tree and the leaf descendants of a node in the second tree. The second variant of the problem allows all rectangles. For the database application, we give a 2-approximation algorithm and prove the problem NP-hard. We give a 2.56-approximation algorithm for second variant of the problem and prove it NP-Hard. To our knowledge these are the first results for the problem of sparsely and exactly representing matrices by weighted rectangles. Joint work with Howard Karloff, Flip Korn, and Yuval Rabani. Contact: Yury Makarychev, yury@tti-c.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://ttic.uchicago.edu/pipermail/colloquium/attachments/20090930/1fb62338/attachment.htm