Constrained bandit algorithms whose regret depends on the realized losses, split into a safety term and a bandit-learning term, with a matching lower bound.
(5) 13 ARXIV PREPRINT - JULY 12, 2025 The result above holds since eSt is a polytope (thus, convex) for any t ∈ [T ]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Data-Dependent Regret Bounds for Constrained MABs
Constrained bandit algorithms whose regret depends on the realized losses, split into a safety term and a bandit-learning term, with a matching lower bound.