We propose an algorithm for stochastic and adversarial multiarmed bandits with switching costs, where the algorithm pays a price λλ every time it switches the arm being played. Our algorithm is based on adaptation of the Tsallis-INF algorithm of Zimmert and Seldin (2021) and requires no prior knowledge of the regime or time horizon. In the oblivious adversarial setting it achieves the minimax optimal regret bound of O((λK)1/3T2/3+KT−−−√)O((λK)1/3T2/3+KT), where TT is the time horizon and KK is the number of arms. In the stochastically constrained adversarial regime, which includes the stochastic regime as a special case, it achieves a regret bound of O((λK)2/3T1/3+lnT)∑i≠i∗Δ−1i)O((λK)2/3T1/3+lnT)∑i≠i∗Δi−1), where ΔiΔi are suboptimality gaps and i∗i∗ is the unique optimal arm. In the special case of λ=0λ=0 (no switching costs), both bounds are minimax optimal within constants. We also explore variants of the problem, where switching cost is allowed to change over time. We provide experimental evaluation showing competitiveness of our algorithm with the relevant baselines in the stochastic, stochastically constrained adversarial, and adversarial regimes with fixed switching cost.
An Algorithm for Stochastic and Adversarial Bandits with Switching Costs
2021-01-01
Article/Chapter (Book)
Electronic Resource
English
DDC: | 629 |
THE LAST WORD One-armed bandits
British Library Online Contents | 2005
Air traffic cop eyes `bandits', too
Engineering Index Backfile | 1967
Garage rip-offs - We expose the time bandits
Online Contents | 1998