Spectral Thompson sampling

TL;DR

SpectralTS is a computationally efficient bandit algorithm leveraging graph smoothness, achieving regret bounds comparable to traditional methods, with demonstrated effectiveness on synthetic and real-world data.

Contribution

We introduce SpectralTS, a spectral bandit algorithm that scales well with large choice sets by exploiting graph structure, providing theoretical regret bounds and empirical validation.

Findings

Regret scales as d*sqrt(T ln N), matching known results.

SpectralTS outperforms traditional algorithms in large-scale settings.

Effective on both synthetic and real-world datasets.

Abstract

Thompson Sampling (TS) has attracted 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.