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 1221–1230 of 1262 papers

TitleStatusHype
Towards the D-Optimal Online Experiment Design for Recommender SelectionCode0
Distributionally Robust Policy Evaluation under General Covariate Shift in Contextual BanditsCode0
When is Off-Policy Evaluation (Reward Modeling) Useful in Contextual Bandits? A Data-Centric PerspectiveCode0
Minimum Empirical Divergence for Sub-Gaussian Linear BanditsCode0
Regret Bounds for Thompson Sampling in Episodic Restless Bandit ProblemsCode0
Mitigating Exposure Bias in Online Learning to Rank Recommendation: A Novel Reward Model for Cascading BanditsCode0
Model-free Reinforcement Learning in Infinite-horizon Average-reward Markov Decision ProcessesCode0
Nonparametric Gaussian Mixture Models for the Multi-Armed BanditCode0
Taming the Monster: A Fast and Simple Algorithm for Contextual BanditsCode0
Two-Stage Neural Contextual Bandits for Personalised News RecommendationCode0
Show:102550
← PrevPage 123 of 127Next →

Benchmark Results

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