When: 14-15 Jan., 2008

Where: Rm: 46-5193

Who: Tomaso Poggio, Lorenzo Rosasco & Thomas Serre

Requirements: None

Synopsis: Understanding the processing of information in our cortex is a significant part of understanding how the brain works and of understanding intelligence itself, arguably one of the greatest problems in science today.  In particular, our visual abilities are computationally amazing and we are still far from imitating them with computers.  Thus, visual cortex may well be a good proxy for the rest of the cortex and indeed for intelligence itself.  But despite enormous progress in the physiology and anatomy of the visual cortex, our understanding of the underlying computations remains fragmentary.We will begin by reviewing research at CBCL on the problem of intelligence.

The Center for Biological & Computational Learning at MIT was founded with the belief that learning is at the very core of the problem of intelligence, both biological and artificial, and is the gateway to understanding how the human brain works and to making intelligent machines. Thus CBCL studies the problem of learning within a multidisciplinary approach. Its main goal is to nurture serious research on the mathematics, the engineering and the neuroscience of learning. We will continue with a brief review of modern learning theory, followed by a more specialized session on current approaches and open questions in learning theory.

In the second day we will briefly review the anatomy and the physiology of primate visual cortex and then describe a class of quantitative models of the ventral stream for object recognition, which, heavily constrained by physiology and biophysics, have been developed during the last two decades and which have been recently shown to be quite successful in explaining several physiological data across different visual areas. We will discuss their performance and architecture from the point of view of state-of-the-art computer vision system. Surprisingly, such models also mimic the level of human performance in difficult rapid image categorization tasks in which human vision is forced to operate in a feedforward mode. In the last session, we will focus on current topics of research on the computational architecture of visual cortex and discuss their implications for advancing computer vision technology.

Last update: Tue 15-Jan-2008

Class 1: Mon 14 Jan, 10:30am-12:00pm

Introduction to learning theory and applications (Tomaso Poggio): slides

Possible additional readings:

Learning theory

F. Cucker and S. Smale. On The Mathematical Foundations of Learning. Bulletin of the American Mathematical Society, 2002.

Poggio, T. and S. Smale. The Mathematics of Learning: Dealing with Data , Notices of the American Mathematical Society (AMS) , Vol. 50, No. 5, 537-544, 2003. (See journal issue at AMS Notices)

Poggio, T., R. Rifkin, S. Mukherjee and P. Niyogi. General Conditions for Predictivity in Learning Theory , Nature , Vol. 428, 419-422, 2004.

Bousquet, O., S. Boucheron and G. Lugosi: Introduction to Statistical Learning Theory. Advanced Lectures on Machine Learning Lecture Notes in Artificial Intelligence 3176, 169-207. (Eds.) Bousquet, O., U. von Luxburg and G. Rätsch, Springer, Heidelberg, Germany, 2004.

Bioinformatics

Pomeroy, S.L., P. Tamayo, M. Gaasenbeek, L.M. Sturia, M. Angelo, M.E. McLaughlin, J.Y.H. Kim, L.C. Goumnerova, P.M. Black, C. Lau, J.C. Allen, D. Zagzag, M.M. Olson, T. Curran, C. Wetmore, J.A. Biegel, T. Poggio, S. Mukherjee, R. Rifkin, A. Califano, G. Stolovitzky, D.N. Louis, J.P. Mesirov, E.S. Lander and T.R. Golub. Prediction of Central Nervous System Embryonal Tumour Outcome Based on Gene Expression , Nature (Letters to Nature) , Vol. 415, 436-442, 2002.

Face detection

B. Heisele, T. Serre and T. Poggio. A component-based framework for face detection and identification. In: International Journal of Computer Vision, to appear , 2007.

Computer Graphics

Ezzat, T., G. Geiger and T. Poggio. Trainable Videorealistic Speech Animation . In: Proceedings of ACM SIGGRAPH 2002, San Antonio, TX, 388-398, 2002.

Useful Links

• MIT 9.520 course: Statistical Learning Theory

Class 2: Mon 14 Jan, 2:00pm-3:30pm

Advanced topics in learning theory (Lorenzo Rosasco): slides

Possible additional readings:

Manifold learning

M. Belkin, P. Niyogi, V. Sindhwani. Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples, 2006.

Class 3: Tue 15 Jan, 10:30pm-12:00pm

Introduction to computational vision (Tomaso Poggio): slides

Models of object recognition in the cortex (Thomas Serre): slides