Top Two Algorithms Revisited
This work addresses a theoretical gap for researchers in multi-armed bandits, offering incremental improvements by extending analysis to non-parametric settings.
The paper tackles the lack of theoretical guarantees for Top Two algorithms in best arm identification beyond Gaussian arms, providing a general analysis that leads to supported algorithms for bounded distributions and shows the leader selection step can be replaced with alternatives like the empirical best arm.
Top Two algorithms arose as an adaptation of Thompson sampling to best arm identification in multi-armed bandit models (Russo, 2016), for parametric families of arms. They select the next arm to sample from by randomizing among two candidate arms, a leader and a challenger. Despite their good empirical performance, theoretical guarantees for fixed-confidence best arm identification have only been obtained when the arms are Gaussian with known variances. In this paper, we provide a general analysis of Top Two methods, which identifies desirable properties of the leader, the challenger, and the (possibly non-parametric) distributions of the arms. As a result, we obtain theoretically supported Top Two algorithms for best arm identification with bounded distributions. Our proof method demonstrates in particular that the sampling step used to select the leader inherited from Thompson sampling can be replaced by other choices, like selecting the empirical best arm.