Communication Dans Un Congrès Année : 2024

Thompson Sampling For Combinatorial Bandits: Polynomial Regret and Mismatched Sampling Paradox

Résumé

We consider Thompson Sampling (TS) for linear combinatorial semi-bandits and subgaussian rewards. We propose the first known TS whose finite-time regret does not scale exponentially with the dimension of the problem. We further show the mismatched sampling paradox: A learner who knows the rewards distributions and samples from the correct posterior distribution can perform exponentially worse than a learner who does not know the rewards and simply samples from a well-chosen Gaussian posterior. The code used to generate the experiments is available at https://github.com/RaymZhang/CTS-Mismatched-Paradox

Mots clés

Fichier principal
Vignette du fichier
NeurIPS-2024-thompson-sampling-for-combinatorial-bandits-polynomial-regret-and-mismatched-sampling-paradox-Paper-Conference.pdf (763.98 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05088046 , version 1 (28-05-2025)

Licence

Identifiants

  • HAL Id : hal-05088046 , version 1

Citer

Raymond Zhang, Richard Combes. Thompson Sampling For Combinatorial Bandits: Polynomial Regret and Mismatched Sampling Paradox. 38. Conference on Neural Information Processing Systems, NeurIPS 2024, Dec 2024, Vancouver, Canada. ⟨hal-05088046⟩
64 Consultations
90 Téléchargements

Partager

  • More