AISTATS 2024poster7 citations

Directional Optimism for Safe Linear Bandits

Spencer Hutchinson, Berkay Turan, Mahnoosh Alizadeh

Abstract

The safe linear bandit problem is a version of the classical stochastic linear bandit problem where the learner’s actions must satisfy an uncertain constraint at all rounds. Due its applicability to many real-world settings, this problem has received considerable attention in recent years. By leveraging a novel approach that we call directional optimism, we find that it is possible to achieve improved regret guarantees for both well-separated problem instances and action sets that are finite star convex sets. Furthermore, we propose a novel algorithm for this setting that improves on existing algorithms in terms of empirical performance, while enjoying matching regret guarantees. Lastly, we introduce a generalization of the safe linear bandit setting where the constraints are convex and adapt our algorithms and analyses to this setting by leveraging a novel convex-analysis based approach.

BibTeX
@InProceedings{pmlr-v238-hutchinson24a,
  title = 	 {Directional Optimism for Safe Linear Bandits},
  author =       {Hutchinson, Spencer and Turan, Berkay and Alizadeh, Mahnoosh},
  booktitle = 	 {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {658--666},
  year = 	 {2024},
  editor = 	 {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen},
  volume = 	 {238},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {02--04 May},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v238/hutchinson24a/hutchinson24a.pdf},
  url = 	 {https://proceedings.mlr.press/v238/hutchinson24a.html},
  abstract = 	 {The safe linear bandit problem is a version of the classical stochastic linear bandit problem where the learner’s actions must satisfy an uncertain constraint at all rounds. Due its applicability to many real-world settings, this problem has received considerable attention in recent years. By leveraging a novel approach that we call directional optimism, we find that it is possible to achieve improved regret guarantees for both well-separated problem instances and action sets that are finite star convex sets. Furthermore, we propose a novel algorithm for this setting that improves on existing algorithms in terms of empirical performance, while enjoying matching regret guarantees. Lastly, we introduce a generalization of the safe linear bandit setting where the constraints are convex and adapt our algorithms and analyses to this setting by leveraging a novel convex-analysis based approach.}
}
Directional Optimism for Safe Linear Bandits · AISTATS 2024