A Note on Zeroth-Order Optimization on the Simplex
Machine Learning
2022-08-03 v1 Computer Science and Game Theory
Optimization and Control
Abstract
We construct a zeroth-order gradient estimator for a smooth function defined on the probability simplex. The proposed estimator queries the simplex only. We prove that projected gradient descent and the exponential weights algorithm, when run with this estimator instead of exact gradients, converge at a rate.
Keywords
Cite
@article{arxiv.2208.01185,
title = {A Note on Zeroth-Order Optimization on the Simplex},
author = {Tijana Zrnic and Eric Mazumdar},
journal= {arXiv preprint arXiv:2208.01185},
year = {2022}
}