Variable Metric Stochastic Approximation Theory
Data Analysis, Statistics and Probability
2009-08-26 v1
Abstract
We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced online versions of the BFGS algorithm and its limited-memory LBFGS variant. We also discuss the implications of our results for learning from expert advice.
Cite
@article{arxiv.0908.3529,
title = {Variable Metric Stochastic Approximation Theory},
author = {Peter Sunehag and Jochen Trumpf and S. V. N. Vishwanathan and Nicol Schraudolph},
journal= {arXiv preprint arXiv:0908.3529},
year = {2009}
}
Comments
Correctment of theorem 3.4. from AISTATS 2009 article