Novembre 2014 - Lisa_teatalk - mailman.iro.umontreal.ca

Schedule of Talks before and after NIPS 2014
by Kyung Hyun Cho 14 Déc '14

14 Déc '14

Dear all, We will have a number of visitors at the lab before and after NIPS 2014. Many of the talks will be interesting to many of the lab members, and I would like to recommend all to be at all of the talks. Also, I'd like to warmly welcome anyone on this mailing list who's not at the LISA lab to attend the talks. Most of the talks will be held at AA3195 (Pavillon André-Aisenstadt, 2920 Chemin de la tour, Montréal). Please, check back frequently for the locations of the talks on http://lisa.iro.umontreal.ca/seminars_en.html. There may be a few more talks added in a next few days, so again, please check the schedule online frequently. Best, - Cho *= Schedule = * + *3 Dec 2014* (Wed) - 13.30 - 14.30 <A Multiplicative Model for Learning Distributed Text-Based Attribute Representations> by Ryan Kiros (University of Toronto) + *4 Dec 2014* (Thu) - 10.30 - 11.30 <What if computer can perceive the world like we do?> by Xinyi Chen, Peng (Eric) Ding (DeepGlint) - 13.30 - 14.30 <Learning Compact Yet Rich Models for Text> by Sarath Chandar (IBM India) + *5 Dec 2014* (Fri) - 10.30 - 11.30 <Joint Training of Deep Autoencoders> by Dr. Yingbo Zhou (University at Buffalo, The State University of New York) - 13.30 - 14.30 <How Watson Learns Superhuman Jeopardy! Strategies> by Dr. Gerald Tesauro (IBM T. J. Watson Research Center) + *15 Dec 2014* (Mon) - 13.30 - 14.30 <TBD> by Prof. Yung-Kyun Noh (KAIST) + *16 Dec 2014* (Tue) - 13.30 - 14.30 <General Stochastic Networks for Classification> by Matthias Zohrer (TU Graz)

2 6

Tea Talk 26 Nov Wed @13.30 AA3195 by Antoine Chassang
by Kyung Hyun Cho 25 Nov '14

25 Nov '14

Dear all, Antoine Chassang will tell us about his work on face detection during his internship at the LISA lab. The talk will be held at the usual place AA3195 starting from 13.30 next Wednesday (26 Nov). Hope to see many of you at the talk! - Cho === - Speaker: Antoine Chassang - Date/Time: 26 Nov (Wed) 13.30 - 14.30 - Place: AA3195 - Title: Face Detection - Abstract: Whereas the state of the art is relying on viewpoint-specialized models which need additional pose-related annotations, we trained convolutional networks on a large dataset (700k faces extracted from Google Images) to learn invariance to pose. We evaluated our method on FDDB, a standard benchmark for face detection, and we show that we can achieve state of the art performance. To speed up the inference, we investigated model compression using a teacher-student formulation where the student network learns to reproduce the teacher output. The compressed model is deeper and thinner, with less parameters than the teacher.

1 1

Talk 5 Dec Fri @13.30 AA3195 by Dr. Gerald Tesauro
by Kyung Hyun Cho 21 Nov '14

21 Nov '14

Dear all, This is an early reminder of the talk by Dr. Gerald Tesauro from IBM Research Center on *5 Dec *(Friday) starting from 13.30 at AA3195. He will tell us about the exciting story behind how Watson beat human competitors on Jeopardy!. Best, - Cho P.S. From now on, you can check the schedules of the future (and past) seminars at http://lisa.iro.umontreal.ca/seminars_en.html. The same page in French will appear soon. === - Speaker: Dr. Gerald Tesauro (IBM TJ Watson Research Center) - Date/Time: 13.30 - 14.30, 5 Dec - Place: AA3195 - Title: How Watson Learns Superhuman Jeopardy! Strategies - Abstract: Major advances in Question Answering technology were needed for Watson to play "Jeopardy!" at championship level -- the show requires rapid-fire answers to challenging natural language questions, broad general knowledge, high precision, and accurate confidence estimates. In addition, Jeopardy! features four types of decision making carrying great strategic importance: (1) selecting the next clue when in control of the board; (2) deciding whether to attempt to buzz in; (3) wagering on Daily Doubles; (4) wagering in Final Jeopardy. This talk describes how Watson makes the above decisions using innovative quantitative methods that, in principle, maximize Watson's overall winning chances. We first describe our development of faithful simulation models of human contestants and the Jeopardy! game environment. We then present specific learning/optimization methods used in each strategy algorithm: these methods span a range of popular AI research topics, including Bayesian inference, game theory, Dynamic Programming, Reinforcement Learning, and real-time "rollouts." Application of these methods yielded superhuman game strategies for Watson that significantly enhanced its overall competitive record. Joint work with David Gondek, Jon Lenchner, James Fan and John Prager. - Speaker Bio: Gerald Tesauro is a Research Staff Member at IBM's TJ Watson Research Center. He is best known for creating TD-Gammon, a self-teaching neural network that learned to play backgammon at human world championship level. He has also worked on theoretical and applied machine learning in a wide variety of other settings, including multi-agent learning, dimensionality reduction, computer virus recognition, computer chess (Deep Blue), intelligent e-commerce agents and autonomic computing. Dr. Tesauro serves on the NIPS Foundation Board of Directors, and was recently elected AAAI Fellow.

1 0

Talk 21 Nov Fri @15.00 AA3195 by Dr. John Hershey
by Kyung Hyun Cho 21 Nov '14

21 Nov '14

Dear all, Dr. John Hershey (Mistubishi Electric Research Labs, US) will tell us about the connection between model-free approaches (e.g. deep neural networks) and model-based approaches (e.g. probabilistic graphical models) and how we can utilize this connection. His talk will start from 15.00 next Friday (21 Nov) at the usual place AA3195. Hope to see many of you there! - Cho === - Speaker: Dr. John Hershey (Mistubishi Electric Research Labs) - Date/Time: 15.00 - 16.00, 21 Nov - Place: AA3195 - Title: Deep Unfolding: Infusing Deep Architectures with Generative Model Inference - Abstract: Model-based methods and deep neural networks have both been tremendously successful paradigms in machine learning. In model-based methods, problem domain knowledge can be built into the constraints of the model, typically at the expense of difficulties during inference. In contrast, deterministic deep neural networks are constructed in such a way that inference is straightforward, but their architectures are rather generic and it can be unclear how to incorporate problem domain knowledge. This work aims to obtain the advantages of both approaches. To do so, we start with a model-based approach and unfold the iterations of its inference method to form a layer-wise structure. We then decouple the model parameters across layers to increase the network's learning capacity. This results in novel neural-network-like architectures that incorporate our model-based constraints, but can be trained discriminatively to perform fast and accurate inference. We show how this framework can be applied to a non-negative matrix factorization model to obtain a new kind of non-negative deep neural network, that can be trained using a multiplicative backpropagation-style update algorithm. We present speech enhancement experiments showing that our approach is competitive with conventional neural networks despite using far fewer parameters. - Bio: John Hershey is a researcher at Mistubishi Electric Research Labs (MERL), in Cambridge, MA since 2010. Prior to that, John spent 5 years as a researcher at IBM's Watson Research Center in New York, in the Speech Algorithms and Engines group, and one year as visiting researcher in the speech group at Microsoft Research, in Redmond, WA. His obtained his Ph.D. at the University of California, San Diego. He is currently working on machine learning for signal enhancement and separation, speech recognition, language processing, and adaptive user interfaces.

1 1

Fwd: ***AUJOURD'HUI*** : COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC / Martin Wainwright
by Yoshua Bengio 20 Nov '14

20 Nov '14

Reminder: one of the highest-profile machine learning researchers is speaking at 4pm in PAA 1360. ---------- Forwarded message ---------- From: <crm(a)crm.umontreal.ca> Date: Thu, Nov 20, 2014 at 9:46 AM Subject: ***AUJOURD'HUI*** : COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC / Martin Wainwright To: activites(a)crm.umontreal.ca ****************************************************************** COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC Montréal ****************************************************************** DATE : Le jeudi 20 novembre 2014 / Thursday, November 20, 2014 HEURE / TIME : 16 h / 4:00 p.m. CONFERENCIER(S) / SPEAKER(S) : Martin Wainwright (University of California, Berkeley) TITRE / TITLE : High-dimensional phenomena in mathematical statistics and convex analysis LIEU / PLACE : CRM, UdeM, Pav. André-Aisenstadt, 2920, ch. de la Tour, salle 1360 RESUME / ABSTRACT : Statistical models in which the ambient dimension is of the same order or larger than the sample size arise frequently in different areas of science and engineering. Although high-dimensional models of this type date back to the work of Kolmogorov, they have been the subject of intensive study over the past decade, and have interesting connections to many branches of mathematics (including concentration of measure, random matrix theory, convex geometry, and information theory). In this talk, we provide a broad overview of the general area, including vignettes on phase transitions in high-dimensional graph recovery, and randomized approximations of convex programs. ****************************************************************** Un café sera servi à 15h30 / Coffee will be served at 3:30 pm ****************************************************************** http://www.crm.umontreal.ca/Colloques/colloqueSMQ-Montreal.html

1 0

Fwd: COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC (20/11/2014, Martin Wainwright)
by Yoshua Bengio 17 Nov '14

17 Nov '14

Martin Wainwright is an outstanding machine learning researcher and he gives a talk Thursday at 4pm room 1360. Don't miss it! ---------- Forwarded message ---------- From: <crm(a)crm.umontreal.ca> Date: Mon, Nov 17, 2014 at 10:32 AM Subject: COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC (20/11/2014, Martin Wainwright) To: activites(a)crm.umontreal.ca ****************************************************************** COLLOQUE DES SCIENCES MATHÉMATIQUES DU QUÉBEC Montréal ****************************************************************** DATE : Le jeudi 20 novembre 2014 / Thursday, November 20, 2014 HEURE / TIME : 16 h / 4:00 p.m. CONFERENCIER(S) / SPEAKER(S) : Martin Wainwright (University of California, Berkeley) TITRE / TITLE : High-dimensional phenomena in mathematical statistics and convex analysis LIEU / PLACE : CRM, UdeM, Pav. André-Aisenstadt, 2920, ch. de la Tour, salle 1360 RESUME / ABSTRACT : Statistical models in which the ambient dimension is of the same order or larger than the sample size arise frequently in different areas of science and engineering. Although high-dimensional models of this type date back to the work of Kolmogorov, they have been the subject of intensive study over the past decade, and have interesting connections to many branches of mathematics (including concentration of measure, random matrix theory, convex geometry, and information theory). In this talk, we provide a broad overview of the general area, including vignettes on phase transitions in high-dimensional graph recovery, and randomized approximations of convex programs. ****************************************************************** Un café sera servi à 15h30 / Coffee will be served at 3:30 pm ****************************************************************** http://www.crm.umontreal.ca/Colloques/colloqueSMQ-Montreal.html

1 0

Tea Talk 12 Nov Wed @15.00 AA3195 by Nicolas Chapados
by Kyung Hyun Cho 11 Nov '14

11 Nov '14

Dear all, Nicolas Chapados (ApSTAT Technology, Ph.D. from LISA Lab) will tell us about his latest work on modelling the time series of integer counts, which was presented at ICML 2014 this year. The talk will be held at the usual place AA3195, but will *start at 15.00*. Hope to see many of you at the talk! - Cho === - Speaker: Dr. Nicolas Chapados, ApSTAT Technology - Date/Time: 12 Nov (Wed) 15.00 - 16.00 - Place: AA3195 - Title: Effective Bayesian Modeling of Groups of Related Count Time Series - Abstract: Time series of integer count data arise in a wide variety of practical situations: the number of clicks on an URL in a given time span, the time-dependent consumer demand for pink T-shirts at a fashion store, or the number of spare parts needed to keep a fleet of fighter jets in working order on an aircraft carrier. When these counts are either small, highly variable or contain a large number of zero values — the signature of rare events — classical time series models are grossly misspecified and often yield unreasonable forecast distributions. In this talk, I describe a hierarchical probabilistic state-space model for count data, for which approximate Bayesian inference is simple and fast. The model can handle arbitrary explanatory variables and recognizes the common scenario of “weak coupling” between a group of related time series (e.g. the demand for a seasonal product at several stores of the same chain should share similar seasonal patterns without necessarily exhibiting strong cross-correlations). I illustrate the forecasting performance on a number of benchmarks from supply chain planning, although the range of applicability of the approach is much wider. I also show that the model can extrapolate useful seasonalities from extremely short time series (only 4 observations!) by automatically borrowing statistical strength from related longer series. This work was presented at the ICML’14 conference in Beijing, last June.

1 1

Tea Talk 5 Nov Wed @13.30 AA3195 by Li Yao
by Kyung Hyun Cho 03 Nov '14

03 Nov '14

Dear all, Li will tell us about exciting recent developments on integrating attention mechanisms to neural networks on Wednesday. The talk will be held at the usual place AA3195 and usual time of 13.30. Hope to see many of you there! === - Speaker: Li Yao - Title: Learning where to look - Abstract: This talk focuses on computational models of "learning where to look". There has been some recent work in this direction using feedforward/recurrent neural networks, Boltzmann machines, neural autoregressive density estimators, and inevitably some reinforcement learning techniques. This talk will start with this paper "Recurrent models of visual attention" (NIPS14). The ideas from other related papers will also be briefly discussed.

1 0