Poster
Ensemble sampling for linear bandits: small ensembles suffice
David Janz · Alexander Litvak · Csaba Szepesvari
[
Abstract
]
Fri 13 Dec 4:30 p.m. PST
— 7:30 p.m. PST
Abstract:
We provide the first useful and rigorous analysis of ensemble sampling for the stochastic linear bandit setting. In particular, we show that, under standard assumptions, for a $d$-dimensional stochastic linear bandit with an interaction horizon $T$, ensemble sampling with an ensemble of size of order $\smash{d \log T}$ incurs regret at most of the order $\smash{(d \log T)^{5/2} \sqrt{T}}$. Ours is the first result in any structured setting not to require the size of the ensemble to scale linearly with $T$---which defeats the purpose of ensemble sampling---while obtaining near $\smash{\sqrt{T}}$ order regret. Ours is also the first result that allows infinite action sets.
Live content is unavailable. Log in and register to view live content