PAC-Bayes Iterated Logarithm Bounds for Martingale Mixtures
Machine Learning
2015-06-23 v1 Probability
Machine Learning
Abstract
We give tight concentration bounds for mixtures of martingales that are simultaneously uniform over (a) mixture distributions, in a PAC-Bayes sense; and (b) all finite times. These bounds are proved in terms of the martingale variance, extending classical Bernstein inequalities, and sharpening and simplifying prior work.
Keywords
Cite
@article{arxiv.1506.06573,
title = {PAC-Bayes Iterated Logarithm Bounds for Martingale Mixtures},
author = {Akshay Balsubramani},
journal= {arXiv preprint arXiv:1506.06573},
year = {2015}
}