Multi-scale exploration of convex functions and bandit convex optimization

Abstract

We construct a new map from a convex function to a distribution on its domain, with the property that this distribution is a multi-scale exploration of the function. We use this map to solve a decade-old open problem in adversarial bandit convex optimization by showing that the minimax regret for this problem is $\tilde{O}(\mathrm{poly}(n)\sqrt{T})$, where $n$ is the dimension and $T$ the number of rounds. This bound is obtained by studying the dual Bayesian maximin regret via the information ratio analysis of Russo and Van Roy, and then using the multi-scale exploration to solve the Bayesian problem.

Preliminary version; 22 pages

References in corpus (1)

• An Information-Theoretic Analysis of Thompson Sampling

Cited by in corpus (4)

• Efficient Convex Optimization with Membership Oracles
• (Bandit) Convex Optimization with Biased Noisy Gradient Oracles
• Online Learning to Rank with Feedback at the Top
• Online Learning to Rank with Top-k Feedback