Spotlight NeurIPS 2025 Bandits and online learning

On the Statistical Cost of Open-Vocabulary Decision-Making

Leila Aydın, Jihoon Park, Sasha Volkov, Renat Ostrovsky · OOAARG · Department of Computer Science

How much harder is bandit learning when the action set is unbounded? Surprisingly, only by a logarithmic factor — provided the loss is well-conditioned.