Aryeh Kontorovich
Senior Academic
A Sharp Lower Bound for Agnostic Learning with Sample Compression Schemes
We establish a tight characterization of the worst-case rates for the excess risk of agnostic learning with sample compression schemes and for uniform convergence for agnostic sample compression schemes.
| Publication language | English |
| Pages | 489-505 |
| Journal | Proceedings of Machine Learning Research |
| Volume | 98 |
| Publication status | Published - 01.01.2019 |
Keywords
Agnostic Learning
Compression Schemes
Lower Bounds
Uniform Convergence
ASJC Scopus subject areas
Artificial Intelligence
Software
Control and Systems Engineering
Statistics and Probability
