all AI news
Tighter PAC-Bayes Generalisation Bounds by Leveraging Example Difficulty. (arXiv:2210.11289v1 [cs.LG])
Oct. 21, 2022, 1:14 a.m. | Felix Biggs, Benjamin Guedj
stat.ML updates on arXiv.org arxiv.org
We introduce a modified version of the excess risk, which can be used to
obtain tighter, fast-rate PAC-Bayesian generalisation bounds. This modified
excess risk leverages information about the relative hardness of data examples
to reduce the variance of its empirical counterpart, tightening the bound. We
combine this with a new bound for $[-1, 1]$-valued (and potentially
non-independent) signed losses, which is more favourable when they empirically
have low variance around $0$. The primary new technical tool is a novel result …
More from arxiv.org / stat.ML updates on arXiv.org
Jobs in AI, ML, Big Data
Senior ML Researcher - 3D Geometry Processing | 3D Shape Generation | 3D Mesh Data
@ Promaton | Europe
Software Engineer, Data Platforms
@ Whatnot | San Francisco, CA, Los Angeles, CA, New York City, Phoenix, AZ, Seattle, WA, Denver, CO
Staff Data Engineer, Data Platform
@ Lilt | Indianapolis
Business Data Analyst - New Division
@ Breakthru Beverage Group | Toronto, ON, Canada
Data Operations Associate
@ iCapital | New York City, United States
Senior Data Scientist, R&D
@ Plusgrade | Toronto, Ontario