Spectral Thompson Sampling - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Spectral Thompson Sampling

Tomáš Kocák
  • Function : Author
  • PersonId : 955512
Michal Valko
Rémi Munos
  • Function : Author
  • PersonId : 836863


Thompson Sampling (TS) has surged a lot of interest due to its good empirical performance, in particular in the computational advertising. Though successful, the tools for its performance analysis appeared only recently. In this paper, we describe and analyze SpectralTS algorithm for a bandit problem, where the payoffs of the choices are smooth given an underlying graph. In this setting, each choice is a node of a graph and the expected payoffs of the neighboring nodes are assumed to be similar. Although the setting has application both in recommender systems and advertising, the traditional algorithms would scale poorly with the number of choices. For that purpose we consider an effective dimension d, which is small in real-world graphs. We deliver the analysis showing that the regret of SpectralTS scales as d\sqrt(T \ln N) with high probability, where T is the time horizon and N is the number of choices. Since a d\sqrt(T \ln N) regret is comparable to the known results, SpectralTS offers a computationally more efficient alternative. We also show that our algorithm is competitive on both synthetic and real-world data.
Fichier principal
Vignette du fichier
kocak2014spectral.pdf (305.71 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00981575 , version 1 (22-04-2014)
hal-00981575 , version 2 (27-07-2014)


  • HAL Id : hal-00981575 , version 2


Tomáš Kocák, Michal Valko, Rémi Munos, Shipra Agrawal. Spectral Thompson Sampling. AAAI Conference on Artificial Intelligence, Jul 2014, Québec City, Canada. ⟨hal-00981575v2⟩
421 View
1027 Download


Gmail Facebook X LinkedIn More