CS Internal Seminar Series: Massi Pontil and Niloy Mitra

Speaker: Massi Pontil, ISML and Niloy Mitra, CMIC
UCL Contact: Melanie Johnson (Visitors from outside UCL please email in advance).
Date/Time: 13 Dec 11, 16:30 - 17:30
Venue: 6.12
Further Information:

Refreshments will be served in the common room afterwards.

Abstract

Massi Pontil
Multi-task Learning: Theory and Practice

ABSTRACT: We discuss the problem of estimating a structured matrix with a large number of elements. A key motivation for this problem occurs in multi-task learning. In this case, the columns of the matrix correspond to the parameters of different regression or classification tasks, and there is structure due to relations between the tasks.

This problem is important in a variety of application areas, ranging from user modelling, to computer vision, to bioinfomatics, to neuroimaging, to mention just a few. We present a general method to learn the tasks' parameters as well as their structure. Our approach is based on solving a convex optimization problem, involving a data term and a penalty term. We highlight different types of penalty terms which are of practical and theoretical importance. They implement structural relations between the tasks and achieve a sparse representations of parameters. We address computational issues as well as the predictive performance of the method.

Niloy Mitra
Of Form and Function

ABSTRACT: The 3D world around us is complex and often cluttered. As acquisition devices improve and become economical, we are being flooded with 3D content in the form of scanned or directly modeled objects. There is an urgent need to analyze and understand such low-level data. At present, there is a fundamental disconnect between such data and our high-level understanding of physical objects. To address this problem, we are working towards jointly analyzing large data collections to infer "function from form", and then use the knowledge to evolve "form from function". We will report both current findings and glimpse of ongoing efforts.