9

Near-Optimal Regret for Efficient Stochastic Combinatorial Semi-Bandits

Zichun Ye
Runqi Wang
Xutong Liu
Shuai Li
Main:7 Pages
5 Figures
Bibliography:2 Pages
11 Tables
Appendix:17 Pages
Abstract

The combinatorial multi-armed bandit (CMAB) is a cornerstone of sequential decision-making framework, dominated by two algorithmic families: UCB-based and adversarial methods such as follow the regularized leader (FTRL) and online mirror descent (OMD). However, prominent UCB-based approaches like CUCB suffer from additional regret factor logT\log T that is detrimental over long horizons, while adversarial methods such as EXP3.M and HYBRID impose significant computational overhead. To resolve this trade-off, we introduce the Combinatorial Minimax Optimal Strategy in the Stochastic setting (CMOSS). CMOSS is a computationally efficient algorithm that achieves an instance-independent regret of O((logk)2kmT)O\big( (\log k)^2\sqrt{kmT}\big ) under semi-bandit feedback, where mm is the number of arms and kk is the maximum cardinality of a feasible action. Crucially, this result eliminates the dependency on logT\log T and matches the established Ω(kmT)\Omega\big( \sqrt{kmT}\big) lower bound up to O((logk)2)O\big((\log k)^2\big). We then extend our analysis to show that CMOSS is also applicable to cascading feedback. Experiments on synthetic and real-world datasets validate that CMOSS consistently outperforms benchmark algorithms in both regret and runtime efficiency.

View on arXiv
Comments on this paper