This is the public, feature-limited version of the conference webpage. After Registration and login please visit the full version.

Workshop: Deep Learning through Information Geometry

Pratik Chaudhari, Alex Alemi, Varun Jog, Dhagash Mehta, Frank Nielsen, Stefano Soatto, Greg Ver Steeg

2020-12-12T09:20:00-08:00 - 2020-12-12T18:30:00-08:00
Abstract: Attempts at understanding deep learning have come from different disciplines, namely physics, statistics, information theory, and machine learning. These lines of investigation have very different modeling assumptions and techniques; it is unclear how their results may be reconciled together. This workshop builds upon the observation that Information Geometry has strong overlaps with these directions and may serve as a means to develop a holistic understanding of deep learning. The workshop program is designed to answer two specific questions. The first question is: how do geometry of the hypothesis class and information-theoretic properties of optimization inform generalization. Good datasets have been a key propeller of the empirical success of deep networks. Our theoretical understanding of data is however poor. The second question the workshop will focus on is: how can we model data and use the understanding of data to improve optimization/generalization in the low-data regime.

Gather.Town link:


To ask questions please use rocketchat, available only upon registration and login.


2020-12-12T09:20:00-08:00 - 2020-12-12T09:30:00-08:00
Opening Remarks
2020-12-12T09:30:00-08:00 - 2020-12-12T10:15:00-08:00
Keynote 1: Ke Sun
Ke Sun
2020-12-12T10:15:00-08:00 - 2020-12-12T10:30:00-08:00
Contributed Talk 1: The Volume of Non-Restricted Boltzmann Machines and Their Double Descent Model Complexity
Prasad Cheema, Mahito Sugiyama
2020-12-12T10:30:00-08:00 - 2020-12-12T10:45:00-08:00
Contributed Talk 2: From em-Projections to Variational Auto-Encoder
Tian Han
2020-12-12T10:45:00-08:00 - 2020-12-12T11:30:00-08:00
Keynote 2: Marco Gori
Marco Gori
2020-12-12T12:30:00-08:00 - 2020-12-12T13:15:00-08:00
Keynote 3: Shun-ichi Amari
Shun-ichi Amari
2020-12-12T13:15:00-08:00 - 2020-12-12T14:00:00-08:00
Keynote 4: Alexander Rakhlin
Alexander Rakhlin
2020-12-12T14:15:00-08:00 - 2020-12-12T14:30:00-08:00
Contributed Talk 3: An Information-Geometric Distance on the Space of Tasks
Yansong Gao
2020-12-12T14:30:00-08:00 - 2020-12-12T15:15:00-08:00
Keynote 5: Gintare Karolina Dziugaite
Gintare Karolina Dziugaite
2020-12-12T15:15:00-08:00 - 2020-12-12T16:00:00-08:00
Keynote 6: Guido Montufar
Guido Montufar
2020-12-12T16:00:00-08:00 - 2020-12-12T16:15:00-08:00
Contributed Talk 4: Annealed Importance Sampling with q-Paths
Rob Brekelmans
2020-12-12T16:30:00-08:00 - 2020-12-12T17:00:00-08:00
Panel Discussion and Closing Remarks
2020-12-12T17:00:00-08:00 - 2020-12-12T18:30:00-08:00
Poster Session