On the Suboptimality of Thompson Sampling in High Dimensions - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

On the Suboptimality of Thompson Sampling in High Dimensions

Raymond Zhang
  • Fonction : Auteur
  • PersonId : 1216374
Richard Combes

Résumé

In this paper we consider Thompson Sampling (TS) for combinatorial semi-bandits. We demonstrate that, perhaps surprisingly, TS is sub-optimal for this problem in the sense that its regret scales exponentially in the ambient dimension, and its minimax regret scales almost linearly. This phenomenon occurs under a wide variety of assumptions including both non-linear and linear reward functions, with Bernoulli distributed rewards and uniform priors. We also show that including a fixed amount of forced exploration to TS does not alleviate the problem. We complement our theoretical results with numerical results and show that in practice TS indeed can perform very poorly in some high dimensional situations.
Fichier principal
Vignette du fichier
NeurIPS-2021-on-the-suboptimality-of-thompson-sampling-in-high-dimensions-Paper.pdf (805.62 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03942542 , version 1 (17-01-2023)

Identifiants

  • HAL Id : hal-03942542 , version 1

Citer

Raymond Zhang, Richard Combes. On the Suboptimality of Thompson Sampling in High Dimensions. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Dec 2021, Virtual Conference, France. ⟨hal-03942542⟩
19 Consultations
15 Téléchargements

Partager

Gmail Facebook X LinkedIn More