DeepAI AI Chat
Log In Sign Up

Boltzmann Exploration Done Right

by   Nicolò Cesa-Bianchi, et al.
Università degli Studi di Milano

Boltzmann exploration is a classic strategy for sequential decision-making under uncertainty, and is one of the most standard tools in Reinforcement Learning (RL). Despite its widespread use, there is virtually no theoretical understanding about the limitations or the actual benefits of this exploration scheme. Does it drive exploration in a meaningful way? Is it prone to misidentifying the optimal actions or spending too much time exploring the suboptimal ones? What is the right tuning for the learning rate? In this paper, we address several of these questions in the classic setup of stochastic multi-armed bandits. One of our main results is showing that the Boltzmann exploration strategy with any monotone learning-rate sequence will induce suboptimal behavior. As a remedy, we offer a simple non-monotone schedule that guarantees near-optimal performance, albeit only when given prior access to key problem parameters that are typically not available in practical situations (like the time horizon T and the suboptimality gap Δ). More importantly, we propose a novel variant that uses different learning rates for different arms, and achieves a distribution-dependent regret bound of order K^2 T/Δ and a distribution-independent bound of order √(KT) K without requiring such prior knowledge. To demonstrate the flexibility of our technique, we also propose a variant that guarantees the same performance bounds even if the rewards are heavy-tailed.


page 1

page 2

page 3

page 4


Optimal Algorithms for Stochastic Multi-Armed Bandits with Heavy Tailed Rewards

In this paper, we consider stochastic multi-armed bandits (MABs) with he...

Corruption Robust Exploration in Episodic Reinforcement Learning

We initiate the study of multi-stage episodic reinforcement learning und...

Almost Boltzmann Exploration

Boltzmann exploration is widely used in reinforcement learning to provid...

Explore no more: Improved high-probability regret bounds for non-stochastic bandits

This work addresses the problem of regret minimization in non-stochastic...

Batched Thompson Sampling

We introduce a novel anytime Batched Thompson sampling policy for multi-...

Efficient Reinforcement Learning via Initial Pure Exploration

In several realistic situations, an interactive learning agent can pract...