Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Benjamin Van Roy

Reinforcement Learning, Bit by Bit


Mar 14, 2021
Xiuyuan Lu, Benjamin Van Roy, Vikranth Dwaracherla, Morteza Ibrahimi, Ian Osband, Zheng Wen


  Access Paper or Ask Questions

Simple Agent, Complex Environment: Efficient Reinforcement Learning with Agent State


Mar 08, 2021
Shi Dong, Benjamin Van Roy, Zhengyuan Zhou


  Access Paper or Ask Questions

A Bit Better? Quantifying Information for Bandit Learning


Feb 18, 2021
Adithya M. Devraj, Benjamin Van Roy, Kuang Xu

* 41 pages, 10 figures, 1 table 

  Access Paper or Ask Questions

Deciding What to Learn: A Rate-Distortion Approach


Jan 15, 2021
Dilip Arumugam, Benjamin Van Roy


  Access Paper or Ask Questions

Randomized Value Functions via Posterior State-Abstraction Sampling


Oct 05, 2020
Dilip Arumugam, Benjamin Van Roy


  Access Paper or Ask Questions

Hypermodels for Exploration


Jun 12, 2020
Vikranth Dwaracherla, Xiuyuan Lu, Morteza Ibrahimi, Ian Osband, Zheng Wen, Benjamin Van Roy

* Published as a conference paper at ICLR 2020 

  Access Paper or Ask Questions

Langevin DQN


Feb 17, 2020
Vikranth Dwaracherla, Benjamin Van Roy

* 5 figures, 14 pages 

  Access Paper or Ask Questions

Provably Efficient Reinforcement Learning with Aggregated States


Dec 13, 2019
Shi Dong, Benjamin Van Roy, Zhengyuan Zhou


  Access Paper or Ask Questions

Information-Theoretic Confidence Bounds for Reinforcement Learning


Nov 21, 2019
Xiuyuan Lu, Benjamin Van Roy


  Access Paper or Ask Questions

Comments on the Du-Kakade-Wang-Yang Lower Bounds


Nov 18, 2019
Benjamin Van Roy, Shi Dong


  Access Paper or Ask Questions

Behaviour Suite for Reinforcement Learning


Aug 13, 2019
Ian Osband, Yotam Doron, Matteo Hessel, John Aslanides, Eren Sezener, Andre Saraiva, Katrina McKinney, Tor Lattimore, Csaba Szepezvari, Satinder Singh, Benjamin Van Roy, Richard Sutton, David Silver, Hado Van Hasselt


  Access Paper or Ask Questions

On the Performance of Thompson Sampling on Logistic Bandits


May 12, 2019
Shi Dong, Tengyu Ma, Benjamin Van Roy

* Accepted for presentation at the Conference on Learning Theory (COLT) 2019 

  Access Paper or Ask Questions

An Information-Theoretic Analysis for Thompson Sampling with Many Actions


Oct 01, 2018
Shi Dong, Benjamin Van Roy


  Access Paper or Ask Questions

Deep Exploration via Randomized Value Functions


Jun 06, 2018
Ian Osband, Benjamin Van Roy, Daniel Russo, Zheng Wen


  Access Paper or Ask Questions

Scalable Coordinated Exploration in Concurrent Reinforcement Learning


May 23, 2018
Maria Dimakopoulou, Ian Osband, Benjamin Van Roy


  Access Paper or Ask Questions

Satisficing in Time-Sensitive Bandit Learning


Mar 07, 2018
Daniel Russo, Benjamin Van Roy

* This submission largely supersedes earlier work in arXiv:1704.09028 

  Access Paper or Ask Questions

Gaussian-Dirichlet Posterior Dominance in Sequential Learning


Feb 09, 2018
Ian Osband, Benjamin Van Roy


  Access Paper or Ask Questions

Coordinated Exploration in Concurrent Reinforcement Learning


Feb 05, 2018
Maria Dimakopoulou, Benjamin Van Roy


  Access Paper or Ask Questions

Ensemble Sampling


Nov 22, 2017
Xiuyuan Lu, Benjamin Van Roy


  Access Paper or Ask Questions

A Tutorial on Thompson Sampling


Nov 19, 2017
Daniel Russo, Benjamin Van Roy, Abbas Kazerouni, Ian Osband, Zheng Wen


  Access Paper or Ask Questions

Learning to Price with Reference Effects


Aug 29, 2017
Abbas Kazerouni, Benjamin Van Roy


  Access Paper or Ask Questions

Learning to Optimize via Information-Directed Sampling


Jul 07, 2017
Daniel Russo, Benjamin Van Roy

* arXiv admin note: substantial text overlap with arXiv:1403.5341 

  Access Paper or Ask Questions

On Optimistic versus Randomized Exploration in Reinforcement Learning


Jun 13, 2017
Ian Osband, Benjamin Van Roy

* Extended abstract for RLDM 2017 

  Access Paper or Ask Questions

Why is Posterior Sampling Better than Optimism for Reinforcement Learning?


Jun 13, 2017
Ian Osband, Benjamin Van Roy


  Access Paper or Ask Questions

Time-Sensitive Bandit Learning and Satisficing Thompson Sampling


Apr 28, 2017
Daniel Russo, David Tse, Benjamin Van Roy


  Access Paper or Ask Questions

Conservative Contextual Linear Bandits


Mar 04, 2017
Abbas Kazerouni, Mohammad Ghavamzadeh, Yasin Abbasi-Yadkori, Benjamin Van Roy


  Access Paper or Ask Questions