Title: Beyond SVMs: Next generation machine learning with the MCM

Speaker: Jayadeva, EE Dept

Abstract: Over the last decade and a half, support vector machines (SVMs) have become the paradigm of choice for most learning applications. The first part of this talk will focus on SVMs and how to use them. SVMs and their variants now provide state-of-the-art results for many applications. Surprisingly, computational learning theory tells us that SVMs provide no guarantee for good generalization, and in fact, can do very poorly at times. It is known that the Vapnik-Chervonenkis or VC dimension of SVMs can be very large or infinite. In short, we do not have algorithms that can provide performance guarantees. In the last few years, new sources of data have emerged, ranging from high dimensional micro-array and bio-informatics data, to very large databases emanating from social networks and telecom service providers. The analysis of such big data sources demands performance guarantees. In this talk, we introduce the Minimal Complexity Machine (MCM), which minimizes an exact bound on the VC dimension. This means that the VC dimension of a MCM classifier can be kept small, and it provides a radically new direction to learning. On a number of benchmark datasets, the MCM generalizes better than SVMs. The MCM typically uses one-third the number of support vectors used by a SVM, indicating that the MCM does indeed learn simpler representations.

Bio: Dr. Jayadeva is currently a Professor in the Department of Electrical Engineering, IIT Delhi. He has been a speaker of the IEEE Computer Society Distinguished Visitor Programme. He spent a sabbatical year as a visiting Researcher at IBM India Research Laboratory, from July 2006. One of his papers in Neurocomputing was listed on the Top25 hotlist; he is a recipient of best paper awards from the IETE Journal of Research, and two other conference papers. He holds a US Patent on A/D conversion, another on assessing pronunciation abilities, and is the co-author of the book “Numerical Optimization and Applications”. He has served on the Steering and Program Committees of several international conferences. His research interests include Swarm Intelligence, Machine Learning, Optimization, and VLSI. Amongst his recent works that has received notable attention is the Twin Support Vector Machine, which currently has around 350 citations.