Informational Confidence Bounds for Self-Normalized Averages and Applications
Statistics Theory
2016-11-17 v1 Statistics Theory
Abstract
We present deviation bounds for self-normalized averages and applications to estimation with a random number of observations. The results rely on a peeling argument in exponential martingale techniques that represents an alternative to the method of mixture. The motivating examples of bandit problems and context tree estimation are detailed.
Cite
@article{arxiv.1309.3376,
title = {Informational Confidence Bounds for Self-Normalized Averages and Applications},
author = {Aurélien Garivier},
journal= {arXiv preprint arXiv:1309.3376},
year = {2016}
}