Timezone: »
We introduce a new convergent variant of Q-learning, called speedy Q-learning, to address the problem of slow convergence in the standard form of the Q-learning algorithm. We prove a PAC bound on the performance of SQL, which shows that for an MDP with n state-action pairs and the discount factor \gamma only T=O\big(\log(n)/(\epsilon^{2}(1-\gamma)^{4})\big) steps are required for the SQL algorithm to converge to an \epsilon-optimal action-value function with high probability. This bound has a better dependency on 1/\epsilon and 1/(1-\gamma), and thus, is tighter than the best available result for Q-learning. Our bound is also superior to the existing results for both model-free and model-based instances of batch Q-value iteration that are considered to be more efficient than the incremental methods like Q-learning.
Author Information
Mohammad Gheshlaghi Azar (Radboud University of Nijmegen)
Remi Munos (Google DeepMind)
Mohammad Ghavamzadeh (Facebook AI Research)
Hilbert J Kappen (Radboud University)
More from the Same Authors
-
2019 : Poster and Coffee Break 1 »
Aaron Sidford · Aditya Mahajan · Alejandro Ribeiro · Alex Lewandowski · Ali H Sayed · Ambuj Tewari · Angelika Steger · Anima Anandkumar · Asier Mujika · Hilbert J Kappen · Bolei Zhou · Byron Boots · Chelsea Finn · Chen-Yu Wei · Chi Jin · Ching-An Cheng · Christina Yu · Clement Gehring · Craig Boutilier · Dahua Lin · Daniel McNamee · Daniel Russo · David Brandfonbrener · Denny Zhou · Devesh Jha · Diego Romeres · Doina Precup · Dominik Thalmeier · Eduard Gorbunov · Elad Hazan · Elena Smirnova · Elvis Dohmatob · Emma Brunskill · Enrique Munoz de Cote · Ethan Waldie · Florian Meier · Florian Schaefer · Ge Liu · Gergely Neu · Haim Kaplan · Hao Sun · Hengshuai Yao · Jalaj Bhandari · James A Preiss · Jayakumar Subramanian · Jiajin Li · Jieping Ye · Jimmy Smith · Joan Bas Serrano · Joan Bruna · John Langford · Jonathan Lee · Jose A. Arjona-Medina · Kaiqing Zhang · Karan Singh · Yuping Luo · Zafarali Ahmed · Zaiwei Chen · Zhaoran Wang · Zhizhong Li · Zhuoran Yang · Ziping Xu · Ziyang Tang · Yi Mao · David Brandfonbrener · Shirli Di-Castro · Riashat Islam · Zuyue Fu · Abhishek Naik · Saurabh Kumar · Benjamin Petit · Angeliki Kamoutsi · Simone Totaro · Arvind Raghunathan · Rui Wu · Donghwan Lee · Dongsheng Ding · Alec Koppel · Hao Sun · Christian Tjandraatmadja · Mahdi Karami · Jincheng Mei · Chenjun Xiao · Junfeng Wen · Zichen Zhang · Ross Goroshin · Mohammad Pezeshki · Jiaqi Zhai · Philip Amortila · Shuo Huang · Mariya Vasileva · El houcine Bergou · Adel Ahmadyan · Haoran Sun · Sheng Zhang · Lukas Gruber · Yuanhao Wang · Tetiana Parshakova -
2019 Workshop: Safety and Robustness in Decision-making »
Mohammad Ghavamzadeh · Shie Mannor · Yisong Yue · Marek Petrik · Yinlam Chow -
2019 Poster: Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies »
Yonathan Efroni · Nadav Merlis · Mohammad Ghavamzadeh · Shie Mannor -
2019 Spotlight: Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies »
Yonathan Efroni · Nadav Merlis · Mohammad Ghavamzadeh · Shie Mannor -
2018 Poster: A Lyapunov-based Approach to Safe Reinforcement Learning »
Yinlam Chow · Ofir Nachum · Edgar Duenez-Guzman · Mohammad Ghavamzadeh -
2018 Poster: A Block Coordinate Ascent Algorithm for Mean-Variance Optimization »
Tengyang Xie · Bo Liu · Yangyang Xu · Mohammad Ghavamzadeh · Yinlam Chow · Daoming Lyu · Daesub Yoon -
2017 Poster: Conservative Contextual Linear Bandits »
Abbas Kazerouni · Mohammad Ghavamzadeh · Yasin Abbasi · Benjamin Van Roy -
2016 : Bert Kappen (Radboud University) »
Hilbert J Kappen -
2016 Poster: Safe Policy Improvement by Minimizing Robust Baseline Regret »
Mohammad Ghavamzadeh · Marek Petrik · Yinlam Chow -
2015 Workshop: Machine Learning for (e-)Commerce »
Esteban Arcaute · Mohammad Ghavamzadeh · Shie Mannor · Georgios Theocharous -
2015 Poster: Policy Gradient for Coherent Risk Measures »
Aviv Tamar · Yinlam Chow · Mohammad Ghavamzadeh · Shie Mannor -
2015 Poster: Black-box optimization of noisy functions with unknown smoothness »
Jean-Bastien Grill · Michal Valko · Remi Munos · Remi Munos -
2014 Workshop: Large-scale reinforcement learning and Markov decision problems »
Benjamin Van Roy · Mohammad Ghavamzadeh · Peter Bartlett · Yasin Abbasi Yadkori · Ambuj Tewari -
2014 Poster: Active Regression by Stratification »
Sivan Sabato · Remi Munos -
2014 Poster: Best-Arm Identification in Linear Bandits »
Marta Soare · Alessandro Lazaric · Remi Munos -
2014 Poster: Bounded Regret for Finite-Armed Structured Bandits »
Tor Lattimore · Remi Munos -
2014 Poster: Efficient learning by implicit exploration in bandit problems with side observations »
Tomáš Kocák · Gergely Neu · Michal Valko · Remi Munos -
2014 Poster: Optimistic Planning in Markov Decision Processes Using a Generative Model »
Balázs Szörényi · Gunnar Kedenburg · Remi Munos -
2014 Poster: Algorithms for CVaR Optimization in MDPs »
Yinlam Chow · Mohammad Ghavamzadeh -
2013 Workshop: Bayesian Optimization in Theory and Practice »
Matthew Hoffman · Jasper Snoek · Nando de Freitas · Michael A Osborne · Ryan Adams · Sebastien Bubeck · Philipp Hennig · Remi Munos · Andreas Krause -
2013 Workshop: Planning with Information Constraints for Control, Reinforcement Learning, Computational Neuroscience, Robotics and Games. »
Hilbert J Kappen · Naftali Tishby · Jan Peters · Evangelos Theodorou · David H Wolpert · Pedro Ortega -
2013 Poster: Actor-Critic Algorithms for Risk-Sensitive MDPs »
Prashanth L.A. · Mohammad Ghavamzadeh -
2013 Poster: Approximate Dynamic Programming Finally Performs Well in the Game of Tetris »
Victor Gabillon · Mohammad Ghavamzadeh · Bruno Scherrer -
2013 Oral: Actor-Critic Algorithms for Risk-Sensitive MDPs »
Prashanth L.A. · Mohammad Ghavamzadeh -
2013 Poster: Thompson Sampling for 1-Dimensional Exponential Family Bandits »
Nathaniel Korda · Emilie Kaufmann · Remi Munos -
2013 Poster: Aggregating Optimistic Planning Trees for Solving Markov Decision Processes »
Gunnar Kedenburg · Raphael Fonteneau · Remi Munos -
2012 Poster: Bandit Algorithms boost Brain Computer Interfaces for motor-task selection of a brain-controlled button »
Joan Fruitet · Alexandra Carpentier · Remi Munos · Maureen Clerc -
2012 Poster: Best Arm Identification: A Unified Approach to Fixed Budget and Fixed Confidence »
Victor Gabillon · Mohammad Ghavamzadeh · Alessandro Lazaric -
2012 Poster: Adaptive Stratified Sampling for Monte-Carlo integration of Differentiable functions »
Alexandra Carpentier · Remi Munos -
2012 Poster: Risk-Aversion in Multi-armed Bandits »
Amir Sani · Alessandro Lazaric · Remi Munos -
2011 Poster: Finite Time Analysis of Stratified Sampling for Monte Carlo »
Alexandra Carpentier · Remi Munos -
2011 Poster: Multi-Bandit Best Arm Identification »
Victor Gabillon · Mohammad Ghavamzadeh · Alessandro Lazaric · Sebastien Bubeck -
2011 Poster: Selecting the State-Representation in Reinforcement Learning »
Odalric-Ambrym Maillard · Remi Munos · Daniil Ryabko -
2011 Poster: Sparse Recovery with Brownian Sensing »
Alexandra Carpentier · Odalric-Ambrym Maillard · Remi Munos -
2011 Session: Spotlight Session 2 »
Remi Munos -
2011 Session: Oral Session 1 »
Remi Munos -
2011 Poster: Optimistic Optimization of Deterministic Functions »
Remi Munos -
2010 Spotlight: LSTD with Random Projections »
Mohammad Ghavamzadeh · Alessandro Lazaric · Odalric-Ambrym Maillard · Remi Munos -
2010 Poster: LSTD with Random Projections »
Mohammad Ghavamzadeh · Alessandro Lazaric · Odalric-Ambrym Maillard · Remi Munos -
2010 Poster: Scrambled Objects for Least-Squares Regression »
Odalric-Ambrym Maillard · Remi Munos -
2010 Poster: Error Propagation for Approximate Policy and Value Iteration »
Amir-massoud Farahmand · Remi Munos · Csaba Szepesvari -
2009 Workshop: Probabilistic Approaches for Control and Robotics »
Marc Deisenroth · Hilbert J Kappen · Emo Todorov · Duy Nguyen-Tuong · Carl Edward Rasmussen · Jan Peters -
2009 Poster: Sensitivity analysis in HMMs with application to likelihood maximization »
Pierre-Arnaud Coquelin · Romain Deguest · Remi Munos -
2009 Poster: Compressed Least-Squares Regression »
Odalric-Ambrym Maillard · Remi Munos -
2008 Workshop: Model Uncertainty and Risk in Reinforcement Learning »
Yaakov Engel · Mohammad Ghavamzadeh · Shie Mannor · Pascal Poupart -
2008 Poster: Online Optimization in X-Armed Bandits »
Sebastien Bubeck · Remi Munos · Gilles Stoltz · Csaba Szepesvari -
2008 Poster: Algorithms for Infinitely Many-Armed Bandits »
Yizao Wang · Jean-Yves Audibert · Remi Munos -
2008 Poster: Bounds on marginal probability distributions »
Joris M Mooij · Hilbert J Kappen -
2008 Poster: Regularized Policy Iteration »
Amir-massoud Farahmand · Mohammad Ghavamzadeh · Csaba Szepesvari · Shie Mannor -
2008 Spotlight: Bounds on marginal probability distributions »
Joris M Mooij · Hilbert J Kappen -
2008 Spotlight: Algorithms for Infinitely Many-Armed Bandits »
Yizao Wang · Jean-Yves Audibert · Remi Munos -
2008 Poster: Self-organization using dynamical synapses »
Vicenç Gómez · Andreas Kaltenbrunner · Vicente López · Hilbert J Kappen -
2008 Poster: Particle Filter-based Policy Gradient in POMDPs »
Pierre-Arnaud Coquelin · Romain Deguest · Remi Munos -
2007 Spotlight: Incremental Natural Actor-Critic Algorithms »
Shalabh Bhatnagar · Richard Sutton · Mohammad Ghavamzadeh · Mark P Lee -
2007 Poster: Incremental Natural Actor-Critic Algorithms »
Shalabh Bhatnagar · Richard Sutton · Mohammad Ghavamzadeh · Mark P Lee -
2007 Poster: Fitted Q-iteration in continuous action-space MDPs »
Remi Munos · András Antos · Csaba Szepesvari -
2006 Poster: Bayesian Policy Gradient Algorithms »
Mohammad Ghavamzadeh · Yaakov Engel -
2006 Spotlight: Bayesian Policy Gradient Algorithms »
Mohammad Ghavamzadeh · Yaakov Engel