English

Continuous-Time Multi-Armed Bandits with Controlled Restarts

Machine Learning 2020-07-02 v1 Optimization and Control Machine Learning

Abstract

Time-constrained decision processes have been ubiquitous in many fundamental applications in physics, biology and computer science. Recently, restart strategies have gained significant attention for boosting the efficiency of time-constrained processes by expediting the completion times. In this work, we investigate the bandit problem with controlled restarts for time-constrained decision processes, and develop provably good learning algorithms. In particular, we consider a bandit setting where each decision takes a random completion time, and yields a random and correlated reward at the end, with unknown values at the time of decision. The goal of the decision-maker is to maximize the expected total reward subject to a time constraint τ\tau. As an additional control, we allow the decision-maker to interrupt an ongoing task and forgo its reward for a potentially more rewarding alternative. For this problem, we develop efficient online learning algorithms with O(log(τ))O(\log(\tau)) and O(τlog(τ))O(\sqrt{\tau\log(\tau)}) regret in a finite and continuous action space of restart strategies, respectively. We demonstrate an applicability of our algorithm by using it to boost the performance of SAT solvers.

Keywords

Cite

@article{arxiv.2007.00081,
  title  = {Continuous-Time Multi-Armed Bandits with Controlled Restarts},
  author = {Semih Cayci and Atilla Eryilmaz and R. Srikant},
  journal= {arXiv preprint arXiv:2007.00081},
  year   = {2020}
}
R2 v1 2026-06-23T16:45:00.309Z