Institute to Host Theoretical Machine Learning Talks “Deep Learning: Alchemy or Science?”

The Institute for Advanced Study (IAS) will host a full day of talks with a distinguished panel of experts to discuss advances in theoretical machine learning, organized by Sanjeev Arora, Visiting Professor in the School of Mathematics. The event will take place at the Institute’s Wolfensohn Hall, beginning at 9:45 a.m. on Friday, February 22, 2019.

“Deep learning has led to dramatic progress on problems of artificial intelligence—recognizing images, driving cars, automatic translation between languages—and triggered a new gold rush in the tech sector,” stated Professor Arora. “Some researchers have raised the concern that the rapid progress has led to loss of rigor and precision.”

How accurate is this perception? And what should the field do to combine rapid innovation with solid science and engineering? These are among those questions to be addressed by a panel which will include the following speakers at the forefront of AI research:

Michael Collins
Vikram S. Pandit Professor of Computer Science (Columbia University)
Research Scientist (Google)

Yann LeCun
Director of AI Research (Facebook)
Founding Director (NYU Center for Data Science)
Silver Professor of Computer Science, Neural Science, and Electrical and Computer Engineering (NYU Courant Institute of Mathematical Sciences)

Joelle Pineau
AI Research Lead (Facebook)
Associate Professor and William Dawson Scholar (McGill University)

Zachary Lipton
Assistant Professor and Affiliate Faculty in the Machine Learning Department and Heinz School of Public Policy (Carnegie Mellon University)

Shai Shalev-Shwartz
Professor at the School of Computer Science and Engineering (Hebrew University of Jerusalem)

The Institute’s Program in Theoretical Machine Learning has been led by Arora since September 2017 supported by a $2 million grant from Eric and Wendy Schmidt. The group is specifically focused on fundamental principles related to how algorithms behave in machines, how they learn, and why they are able to make desired predictions and decisions. The program began as a natural extension of existing activities in the Computer Science and Discrete Mathematics (CSDM) program. The CSDM program, which is in its 19th year, holds two weekly seminars, and facilitates intensive research into the practical use of computers and, simultaneously, the unexpected mathematical depth of the abstract notion of “computation” that contributes to the diverse and evolving field of theoretical computer science. CSDM is led by Avi Wigderson, Herbert H. Maass Professor in the School of Mathematics. Next year, fifteen to twenty Members will join the School’s special year program “Optimization, Statistics, and Theoretical Machine Learning” to develop new models, modes of analysis, and novel algorithms.

Support for this event is provided by a grant from the Schwab Charitable Fund made possible by the generosity of Eric and Wendy Schmidt.

Registration is free but is required to participate: https://www.ias.edu/events/deep-learning-2019. The talk will be livestreamed at www.ias.edu/livestream. For more information, please visit http://www.ias.edu/math/tml/dlasagenda. For press interested in interviews or attending, please contact Lee Sandberg at lsandberg@ias.edu.