A parameter-free hedging algorithm

TL;DR

This paper introduces a new parameter-free algorithm for decision-theoretic online learning that performs well with many actions without needing to tune learning rates, addressing a key practical challenge.

Contribution

It proposes a novel, parameter-free algorithm for DTOL that adapts to large action sets and introduces a new regret measure suited for practical applications.

Findings

Achieves competitive regret bounds without parameter tuning

Performs well with a large number of actions

Matches the performance of tuned algorithms under previous regret measures

Abstract

We study the problem of decision-theoretic online learning (DTOL). Motivated by practical applications, we focus on DTOL when the number of actions is very large. Previous algorithms for learning in this framework have a tunable learning rate parameter, and a barrier to using online-learning in practical applications is that it is not understood how to set this parameter optimally, particularly when the number of actions is large. In this paper, we offer a clean solution by proposing a novel and completely parameter-free algorithm for DTOL. We introduce a new notion of regret, which is more natural for applications with a large number of actions. We show that our algorithm achieves good performance with respect to this new notion of regret; in addition, it also achieves performance close to that of the best bounds achieved by previous algorithms with optimally-tuned parameters, according to previous notions of regret.