Relative Deviation Learning Bounds and Generalization with Unbounded Loss Functions
Machine Learning
2016-04-06 v4
Abstract
We present an extensive analysis of relative deviation bounds, including detailed proofs of two-sided inequalities and their implications. We also give detailed proofs of two-sided generalization bounds that hold in the general case of unbounded loss functions, under the assumption that a moment of the loss is bounded. These bounds are useful in the analysis of importance weighting and other learning tasks such as unbounded regression.
Keywords
Cite
@article{arxiv.1310.5796,
title = {Relative Deviation Learning Bounds and Generalization with Unbounded Loss Functions},
author = {Corinna Cortes and Spencer Greenberg and Mehryar Mohri},
journal= {arXiv preprint arXiv:1310.5796},
year = {2016}
}