ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1803.06971
18
115

What Doubling Tricks Can and Can't Do for Multi-Armed Bandits

19 March 2018
Lilian Besson
E. Kaufmann
ArXivPDFHTML
Abstract

An online reinforcement learning algorithm is anytime if it does not need to know in advance the horizon T of the experiment. A well-known technique to obtain an anytime algorithm from any non-anytime algorithm is the "Doubling Trick". In the context of adversarial or stochastic multi-armed bandits, the performance of an algorithm is measured by its regret, and we study two families of sequences of growing horizons (geometric and exponential) to generalize previously known results that certain doubling tricks can be used to conserve certain regret bounds. In a broad setting, we prove that a geometric doubling trick can be used to conserve (minimax) bounds in R_T=O(T)R\_T = O(\sqrt{T})R_T=O(T​) but cannot conserve (distribution-dependent) bounds in R_T=O(log⁡T)R\_T = O(\log T)R_T=O(logT). We give insights as to why exponential doubling tricks may be better, as they conserve bounds in R_T=O(log⁡T)R\_T = O(\log T)R_T=O(logT), and are close to conserving bounds in R_T=O(T)R\_T = O(\sqrt{T})R_T=O(T​).

View on arXiv
Comments on this paper