3pm - 4pm
Wednesday 22 May 2019
A partial overview of the recent trends in the mathematical analysis of deep learning
Speaker: Stephane Chretien (NPL)
Abstract: Deep learning is a new field at the intersection of computer science and mathematics whose main goal is to solve the problem of classifying objects given some training data (and many generalisations of this problem). Despite the extensive activity around devising new algorithms that recently lead to fascinating practical breakthroughs, a complete understanding of the mathematical underpinnings is still lacking. In this talk, I will give an overview of the recent and exciting trends in the mathematics of deep learning and summarise the main open problems in the field.
University of Surrey
This event has passed