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 Csaba Szepesvari

Leveraging Non-uniformity in First-order Non-convex Optimization


May 13, 2021
Jincheng Mei, Yue Gao, Bo Dai, Csaba Szepesvari, Dale Schuurmans

* 48 pages, 10 figures. Accepted at ICML 2021 

  Access Paper or Ask Questions

On the Optimality of Batch Policy Optimization Algorithms


Apr 06, 2021
Chenjun Xiao, Yifan Wu, Tor Lattimore, Bo Dai, Jincheng Mei, Lihong Li, Csaba Szepesvari, Dale Schuurmans

* 29 pages, 8 figures 

  Access Paper or Ask Questions

Improved Regret Bound and Experience Replay in Regularized Policy Iteration


Feb 25, 2021
Nevena Lazic, Dong Yin, Yasin Abbasi-Yadkori, Csaba Szepesvari


  Access Paper or Ask Questions

On the Convergence and Sample Efficiency of Variance-Reduced Policy Gradient Method


Feb 17, 2021
Junyu Zhang, Chengzhuo Ni, Zheng Yu, Csaba Szepesvari, Mengdi Wang


  Access Paper or Ask Questions

Meta-Thompson Sampling


Feb 11, 2021
Branislav Kveton, Mikhail Konobeev, Manzil Zaheer, Chih-wei Hsu, Martin Mladenov, Craig Boutilier, Csaba Szepesvari


  Access Paper or Ask Questions

Nearly Minimax Optimal Reinforcement Learning for Linear Mixture Markov Decision Processes


Jan 07, 2021
Dongruo Zhou, Quanquan Gu, Csaba Szepesvari

* 59 pages, 1 figure 

  Access Paper or Ask Questions

Variational Policy Gradient Method for Reinforcement Learning with General Utilities


Jul 04, 2020
Junyu Zhang, Alec Koppel, Amrit Singh Bedi, Csaba Szepesvari, Mengdi Wang


  Access Paper or Ask Questions

PAC-Bayes Analysis Beyond the Usual Bounds


Jun 23, 2020
Omar Rivasplata, Ilja Kuzborskij, Csaba Szepesvari, John Shawe-Taylor

* Enhanced version of the paper with the same title presented at the NeurIPS 2019 Workshop on Machine Learning with Guarantees. Important update: the PAC-Bayes bound for unbounded loss functions (Section 2.3) is new 

  Access Paper or Ask Questions

Differentiable Meta-Learning in Contextual Bandits


Jun 09, 2020
Branislav Kveton, Martin Mladenov, Chih-Wei Hsu, Manzil Zaheer, Csaba Szepesvari, Craig Boutilier


  Access Paper or Ask Questions

Model-Based Reinforcement Learning with Value-Targeted Regression


Jun 01, 2020
Alex Ayoub, Zeyu Jia, Csaba Szepesvari, Mengdi Wang, Lin F. Yang


  Access Paper or Ask Questions

On the Global Convergence Rates of Softmax Policy Gradient Methods


May 13, 2020
Jincheng Mei, Chenjun Xiao, Csaba Szepesvari, Dale Schuurmans

* 57 pages 

  Access Paper or Ask Questions

Provably Efficient Adaptive Approximate Policy Iteration


Mar 15, 2020
Botao Hao, Nevena Lazic, Yasin Abbasi-Yadkori, Pooria Joulani, Csaba Szepesvari


  Access Paper or Ask Questions

Model Selection in Contextual Stochastic Bandit Problems


Mar 03, 2020
Aldo Pacchiano, My Phan, Yasin Abbasi-Yadkori, Anup Rao, Julian Zimmert, Tor Lattimore, Csaba Szepesvari

* 12 main pages, 2 figures, 14 appendix pages 

  Access Paper or Ask Questions

Differentiable Bandit Exploration


Feb 17, 2020
Craig Boutilier, Chih-Wei Hsu, Branislav Kveton, Martin Mladenov, Csaba Szepesvari, Manzil Zaheer


  Access Paper or Ask Questions

Learning with Good Feature Representations in Bandits and in RL with a Generative Model


Nov 18, 2019
Tor Lattimore, Csaba Szepesvari

* 11 pages 

  Access Paper or Ask Questions

Autonomous exploration for navigating in non-stationary CMPs


Oct 18, 2019
Pratik Gajane, Ronald Ortner, Peter Auer, Csaba Szepesvari


  Access Paper or Ask Questions

Adaptive Exploration in Linear Contextual Bandit


Oct 15, 2019
Botao Hao, Tor Lattimore, Csaba Szepesvari


  Access Paper or Ask Questions

PAC-Bayes with Backprop


Oct 04, 2019
Omar Rivasplata, Vikram M Tankasali, Csaba Szepesvari


  Access Paper or Ask Questions

Exploration-Enhanced POLITEX


Aug 27, 2019
Yasin Abbasi-Yadkori, Nevena Lazic, Csaba Szepesvari, Gellert Weisz


  Access Paper or Ask Questions

Exploration by Optimisation in Partial Monitoring


Jul 24, 2019
Tor Lattimore, Csaba Szepesvari

* simplified algorithm also works for globally observable, bandit and full information games 

  Access Paper or Ask Questions

Randomized Exploration in Generalized Linear Bandits


Jun 21, 2019
Branislav Kveton, Manzil Zaheer, Csaba Szepesvari, Lihong Li, Mohammad Ghavamzadeh, Craig Boutilier


  Access Paper or Ask Questions

Gradient Descent for Sparse Rank-One Matrix Completion for Crowd-Sourced Aggregation of Sparsely Interacting Workers


Apr 25, 2019
Yao Ma, Alex Olshevsky, Venkatesh Saligrama, Csaba Szepesvari


  Access Paper or Ask Questions

Empirical Bayes Regret Minimization


Apr 04, 2019
Chih-Wei Hsu, Branislav Kveton, Ofer Meshi, Martin Mladenov, Csaba Szepesvari


  Access Paper or Ask Questions

Perturbed-History Exploration in Stochastic Linear Bandits


Mar 21, 2019
Branislav Kveton, Csaba Szepesvari, Mohammad Ghavamzadeh, Craig Boutilier


  Access Paper or Ask Questions

An Exponential Efron-Stein Inequality for Lq Stable Learning Rules


Mar 12, 2019
Karim Abou-Moustafa, Csaba Szepesvari

* PMLR Vol. 98, 2019 
* To appear in ALT 2019. arXiv admin note: text overlap with arXiv:1706.05801 

  Access Paper or Ask Questions