Duality Principles for Modern Machine Learning
The ICML Duality Principles workshop brings together researchers working on various duality concepts from many different fields to discuss new applications for modern machine learning, especially focusing on topics such as model understanding, explanation, and adaptation in deep learning and reinforcement learning.
Duality is a pervasive and important principle in mathematics. Not only has it fascinated researchers in many different fields but it has also been used extensively in optimization, statistics, and machine-learning, giving rise to powerful tools such as
- Fenchel duality in convex optimization,
- Representer theorems in kernel methods and Bayesian nonparametrics,
- Dually-flat spaces in information geometry.
Duality played an important role in the past, but lately we do not see much work on duality principles, especially in deep learning. For example, Lagrange duality can be useful for model explanation because it allows us to measure sensitivity of certain perturbations, but this is not yet fully exploited. This slowdown is perhaps due to a growing focus on nonconvex and nonlinear problems where duality does not seem to be directly applicable.
With this workshop, we aim to revive the interest of the ML community in duality principles.
See our Call for Papers here.
This workshop is partially supported by the Bayes-Duality Project, JST CREST Grant Number JPMJCR2112: “A new Bayes-duality principle for adaptive, robust, and lifelong learning of AI systems”.
Announcements
Aug 1, 2023 | Slides for all talks are available in the schedule. |
---|---|
Aug 1, 2023 | The workshop was a great success! Thanks again to all the participants, volunteers, speakers and advisors. We are looking into organizing a 2nd workshop on duality in modern ML next year in 2024! |
Jul 3, 2023 | Contributed talks are announced, see our preliminary schedule. |
Jun 21, 2023 | Paper decisions are out! The list of accepted papers is here. |
May 22, 2023 | The submission deadline is extended to May 29th, anywhere on earth. . |
Apr 6, 2023 | The workshop is open for submissions, see our Call for Papers. |
Mar 20, 2023 | Our workshop proposal has been accepted at ICML 2023! Stay tuned for more updates |
Confirmed Speakers
Amy Zhang Assistant Professor, UT Austin |
Ronny Bergmann Associate Professor, NTNU |
Taiji Suzuki Associate Professor, University of Tokyo |
Ehsan Amid Research Scientist, Google Brain |
Jia-Jie Zhu Research Group Leader, Weierstrass Institute |
Len Spek Post-doc, University of Twente |
Organizers
Zelda Mariet Research Scientist, Google Brain |
Mathieu Blondel Research Scientist, Google Brain |
Thomas Möllenhoff Research Scientist, RIKEN AIP |
Emtiyaz Khan Team Leader, RIKEN AIP |
Volunteers / Logistics
Peter Nickl Research Assistant, RIKEN AIP |
Rob Brekelmans Postdoctoral Fellow, Vector Institute |
Advisory Committee
Suvrit Sra Professor MIT |
Francis Bach Researcher INRIA / ENS |
Nihat Ay Professor TU Hamburg |