English

Rising Multi-Armed Bandits with Known Horizons

Machine Learning 2026-02-16 v2

Abstract

The Rising Multi-Armed Bandit (RMAB) framework models environments where expected rewards of arms increase with plays, which models practical scenarios where performance of each option improves with the repeated usage, such as in robotics and hyperparameter tuning. For instance, in hyperparameter tuning, the validation accuracy of a model configuration (arm) typically increases with each training epoch. A defining characteristic of RMAB is em horizon-dependent optimality: unlike standard settings, the optimal strategy here shifts dramatically depending on the available budget TT. This implies that knowledge of TT yields significantly greater utility in RMAB, empowering the learner to align its decision-making with this shifting optimality. However, the horizon-aware setting remains underexplored. To address this, we propose a novel CUmulative Reward Estimation UCB (CURE-UCB) that explicitly integrates the horizon. We provide a rigorous analysis establishing a new regret upper bound and prove that our method strictly outperforms horizon-agnostic strategies in structured environments like ``linear-then-flat'' instances. Extensive experiments demonstrate its significant superiority over baselines.

Keywords

Cite

@article{arxiv.2602.10727,
  title  = {Rising Multi-Armed Bandits with Known Horizons},
  author = {Seockbean Song and Chenyu Gan and Youngsik Yoon and Siwei Wang and Wei Chen and Jungseul Ok},
  journal= {arXiv preprint arXiv:2602.10727},
  year   = {2026}
}