Replicable UCB-based bandit algorithms are proposed for MAB, linear, and generalized linear bandits, improving prior linear-bandit regret by a factor of O(d/ρ) via a novel replicable ridge regression estimator.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Replicable Bandits with UCB based Exploration
Replicable UCB-based bandit algorithms are proposed for MAB, linear, and generalized linear bandits, improving prior linear-bandit regret by a factor of O(d/ρ) via a novel replicable ridge regression estimator.