AI & Fundamentals

Learning Novel Classes and Attributes While Wandering Within a World - Richard Zemel, Professor, University of Toronto

Please register for this event here.

Abstract:

Real world learning scenarios involve a nonstationary distribution of classes with sequential dependencies among the samples, in contrast to the standard machine learning formulation of drawing samples independently from a fixed, typically uniform distribution. Furthermore, real world interactions demand learning and evaluation on-the-fly, where new classes are learned online, from few or no class labels. I will describe benchmark datasets and models we have developed towards this aim. I will also discuss recent work in which the learning can be formulated in terms of new attributes rather than novel classes. This formulation highlights the value of unsupervised learning, and also enables informative estimates of a model's generalization ability.

Bio:

Richard Zemel is a Professor in the Computer Science Department at Columbia University, on leave from the University of Toronto. He was the Co-Founder and Research Director of the Vector Institute for Artificial Intelligence. His awards include an NVIDIA Pioneers of AI Award, an ONR Young Investigator Award, and a CIFAR AI Chair. He is a Fellow of the Canadian Institute for Advanced Research and is on the Advisory Board of the Neural Information Processing Society.

His research contributions include foundational work on systems that learn useful representations of data with little or no supervision; robust and fair learning algorithms; and machine learning systems for automatic captioning and answering questions about images. He developed the Toronto Paper Matching System, a a system for matching paper submissions to reviewers, which is being used by many conferences. His research is supported by grants from NSERC, CIFAR, Microsoft, Google, Samsung, DARPA and iARPA.

Please register for this event here.