all AI news
SAM as an Optimal Relaxation of Bayes. (arXiv:2210.01620v1 [cs.LG])
Oct. 5, 2022, 1:11 a.m. | Thomas Möllenhoff, Mohammad Emtiyaz Khan
cs.LG updates on arXiv.org arxiv.org
Sharpness-aware minimization (SAM) and related adversarial deep-learning
methods can drastically improve generalization, but their underlying mechanisms
are not yet fully understood. Here, we establish SAM as a relaxation of the
Bayes objective where the expected negative-loss is replaced by the optimal
convex lower bound, obtained by using the so-called Fenchel biconjugate. The
connection enables a new Adam-like extension of SAM to automatically obtain
reasonable uncertainty estimates, while sometimes also improving its accuracy.
By connecting adversarial and Bayesian methods, our work …
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
Senior ML Researcher - 3D Geometry Processing | 3D Shape Generation | 3D Mesh Data
@ Promaton | Europe
Research Assistant/Associate, Health Data Science [LKCMedicine]
@ Nanyang Technological University | NTU Novena Campus, Singapore
Senior Machine Learning Engineer, Portfolio ML
@ Affirm | Remote Canada
[Sessional Lecturer] Foundations of Data Analytics and Machine Learning - APS1070
@ University of Toronto | Toronto, ON, CA
Senior Data Scientist
@ Prosper | United States
Data Analyst
@ ZF Friedrichshafen AG | Coimbatore, TN, IN, 641659