|Seminar on Theoretical Machine Learning|
|Topic:||Learning with little data|
|Affiliation:||University of Toronto; Visitor, School of Mathematics|
|Date:||Monday, December 11|
|Time/Room:||12:30pm - 1:45pm/White-Levy Room|
The current successes of deep neural networks have largely come on classification problems, based on datasets containing hundreds of examples from each category. Humans can easily learn new words or classes of visual objects from very few examples. A fundamental question is how to adapt learning systems to accommodate new classes not seen in training, given only a few examples of each of these classes. I will discuss recent advances in this area, and present ongoing work by my group on various aspects of this problem.