SOTAVerified

Multi-Armed Bandits

Multi-armed bandits refer to a task where a fixed amount of resources must be allocated between competing resources that maximizes expected gain. Typically these problems involve an exploration/exploitation trade-off.

( Image credit: Microsoft Research )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 571–580 of 1262 papers

TitleStatusHype
From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses—0
From Bandits to Experts: On the Value of Side-Observations—0
Improved Algorithms for Adversarial Bandits with Unbounded Losses—0
Improved Algorithms for Misspecified Linear Markov Decision Processes—0
Improved Algorithms for Multi-period Multi-class Packing Problems with Bandit Feedback—0
Improved Best-of-Both-Worlds Guarantees for Multi-Armed Bandits: FTRL with General Regularizers and Multiple Optimal Arms—0
Improved High-Probability Regret for Adversarial Bandits with Time-Varying Feedback Graphs—0
Improved Offline Contextual Bandits with Second-Order Bounds: Betting and Freezing—0
A Tractable Online Learning Algorithm for the Multinomial Logit Contextual Bandit—0
Confidence-Budget Matching for Sequential Budgeted Learning—0
Show:102550
← PrevPage 58 of 127Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1NeuralLinear FullPosterior-MRCumulative regret1.92—Unverified
2Linear FullPosterior-MRCumulative regret1.82—Unverified