Black-Box Reductions for Parameter-free Online Learning in Banach Spaces
Machine Learning
2018-06-27 v2 Optimization and Control
Machine Learning
Abstract
We introduce several new black-box reductions that significantly improve the design of adaptive and parameter-free online learning algorithms by simplifying analysis, improving regret guarantees, and sometimes even improving runtime. We reduce parameter-free online learning to online exp-concave optimization, we reduce optimization in a Banach space to one-dimensional optimization, and we reduce optimization over a constrained domain to unconstrained optimization. All of our reductions run as fast as online gradient descent. We use our new techniques to improve upon the previously best regret bounds for parameter-free learning, and do so for arbitrary norms.
Keywords
Cite
@article{arxiv.1802.06293,
title = {Black-Box Reductions for Parameter-free Online Learning in Banach Spaces},
author = {Ashok Cutkosky and Francesco Orabona},
journal= {arXiv preprint arXiv:1802.06293},
year = {2018}
}
Comments
Appears in Conference on Learning Theory 2018