SOTAVerified

Thompson Sampling

Thompson sampling, named after William R. Thompson, is a heuristic for choosing actions that addresses the exploration-exploitation dilemma in the multi-armed bandit problem. It consists of choosing the action that maximizes the expected reward with respect to a randomly drawn belief.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 176–200 of 655 papers

TitleStatusHype
Convolutional Monte Carlo Rollouts in Go—0
Cost Aware Asynchronous Multi-Agent Active Search—0
Cost-efficient Knowledge-based Question Answering with Large Language Models—0
Asymptotically Optimal Bandits under Weighted Information—0
Counterfactual Data-Fusion for Online Reinforcement Learners—0
Counterfactual Inference under Thompson Sampling—0
Towards Efficient and Optimal Covariance-Adaptive Algorithms for Combinatorial Semi-Bandits—0
Cover Tree Bayesian Reinforcement Learning—0
Customized Nonlinear Bandits for Online Response Selection in Neural Conversation Models—0
Asymptotic Convergence of Thompson Sampling—0
Debiasing Samples from Online Learning Using Bootstrap—0
Decentralized Multi-Agent Active Search and Tracking when Targets Outnumber Agents—0
Deciding What to Learn: A Rate-Distortion Approach—0
Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicine—0
Deep Active Ensemble Sampling For Image Classification—0
Bayesian Quantile and Expectile Optimisation—0
An Information-Theoretic Analysis of Thompson Sampling for Logistic Bandits—0
Deep Contextual Multi-armed Bandits—0
Deep Exploration for Recommendation Systems—0
Deep Hierarchy in Bandits—0
Delay-Adaptive Learning in Generalized Linear Contextual Bandits—0
Adaptively Optimize Content Recommendation Using Multi Armed Bandit Algorithms in E-commerce—0
Differentially Private Federated Bayesian Optimization with Distributed Exploration—0
Diffusion Approximations for Thompson Sampling—0
A Copula approach for hyperparameter transfer learning—0
Show:102550
← PrevPage 8 of 27Next →

No leaderboard results yet.