Deep learning has proved very successful at a host of AI/ML tasks recently, but theoretical understanding of this technique has lagged. The talk will survey ongoing efforts to understand the success of this method, including optimization aspects and a magical ability of vast nets to not overfit on tiny data sets. The talk will be self-contained.
Sanjeev Arora is Charles C. Fitzmorris Professor of Computer Science at Princeton University and Visiting Professor in Mathematics at the Institute for Advanced Study. He works on theoretical computer science and theoretical machine learning. He has received the Packard Fellowship (1997), Simons Investigator Award (2012), G?del Prize (2001 and 2010), ACM Prize in Computing (2012), and the Fulkerson Prize in Discrete Math (2012). He is a fellow of the American Academy of Arts and Sciences and member of the National Academy of Science and was a plenary speaker at the International Congress of Mathematicians in 2018.