機械学習国際会議 ICML

International Conference on Machine Learningで発表された強化学習に関する論文．
（最近のものから順次追加しており，完全なリストではありません．）

ファイナンス †

• Reinforcement learning for optimized trade execution
  Yuriy Nevmyvaka, Yi Feng, Michael Kearns
  ICML 2006, pp. 673-680.

ゲーム †

• Sample-based learning and search with permanent and transient memories
  David Silver, Richard Sutton, and Martin Mueller
  ICML 2008, pp. 968-975.
• Learning algorithms for online principal-agent problems (and selling goods online)
  Vincent Conitzer, Nikesh Garera
  ICML 2006, pp. 209-216.
• Learning to compete, compromise, and cooperate in repeated general-sum games
  Jacob W. Crandall, Michael A. Goodrich
  ICML 2005, pp. 161-168.

ロボット †

• Learning complex motions by sequencing simpler motion templates
  Gerhard Neumann, Wolfgang Maass and Jan Peters
  ICML 2009, pp. 753-760.
• Reinforcement learning by reward-weighted regression for operational space control
  Jan Peters, Stefan Schaal
  ICML 2007, pp. 745-750.

マルチエージェント †

• Dynamic analysis of multiagent Q-learning with ε-greedy exploration
  Eduardo Rodrigues Gomes and Ryszard Kowalczyk
  ICML 2009, pp. 369-376.
• Privacy-preserving reinforcement learning
  Jun Sakuma, Shigenobu Kobayashi, and Rebecca Wright
  ICML 2008, pp. 864-871.
• Conditional random fields for multi-agent reinforcement learning
  Xinhua Zhang, Douglas Aberdeen, S. V. N. Vishwanathan
  ICML 2007, pp. 1143-1150.

階層型強化学習 †

• Hierarchical model-based reinforcement learning: R-max + MAXQ
  Nicholas Jong and Peter Stone
  ICML 2008, pp. 432-439.

多目的強化学習 †

• Learning All Optimal Policies with Multiple Criteria
  Leon Barrett and Srinivas Narayanan
  ICML 2008, pp. 41-47.
• Multi-task reinforcement learning: a hierarchical Bayesian approach
  Aaron Wilson, Alan Fern, Soumya Ray, Prasad Tadepalli
  ICML 2007, pp. 1015-1022.
• Dynamic preferences in multi-criteria reinforcement learning
  Sriraam Natarajan, Prasad Tadepalli
  ICML 2005, pp. 601-608.

転移学習 †

• Transfer of samples in batch reinforcement learning
  Alessandro Lazaric, Marcello Restelli, and Andrea Bonarini
  ICML 2008, pp. 544-551.
• Automatic discovery and transfer of MAXQ hierarchies
  Neville Mehta, Soumya Ray, Prasad Tadepalli, and Thomas Dietterich
  ICML 2008, pp. 648-655.
• Automatic shaping and decomposition of reward functions
  Bhaskara Marthi
  ICML 2007, pp. 601-608.
• Cross-domain transfer for reinforcement learning
  Matthew E. Taylor, Peter Stone
  ICML 2007, pp. 879-886.
• Autonomous shaping: knowledge transfer in reinforcement learning
  George Konidaris, Andrew Barto
  ICML 2006, pp. 489-496.
• Identifying useful subgoals in reinforcement learning by local graph partitioning
  Özgür Şimşek, Alicia P. Wolfe, Andrew G. Barto
  ICML 2005, pp. 816-823.

関係強化学習 †

• Relational temporal difference learning
  Nima Asgharbeygi, David Stracuzzi, Pat Langley
  ICML 2006, pp. 49-56.
• Learning the structure of Factored Markov Decision Processes in reinforcement learning problems
  Thomas Degris, Olivier Sigaud, Pierre-Henri Wuillemin
  ICML 2006, pp. 257-264.

能動学習 †

• Active reinforcement learning
  Arkady Epshteyn, Adam Vogel, and Gerald DeJong
  ICML 2008, pp. 296-303.
• Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs
  Finale Doshi, Joelle Pineau, and Nicholas Roy
  ICML 2008, pp. 256-263.

POMDP †

• Predictive representations for policy gradient in POMDPs
  Abdeslam Boularias and Brahim Chaib-draa
  ICML 2009, pp. 65-72.
• Region-based value iteration for partially observable Markov decision processes
  Hui Li, Xuejun Liao, Lawrence Carin
  ICML 2006, pp. 561-568.

PSR †

• Efficiently learning linear-linear exponential family predictive representations of state
  David Wingate and Satinder Singh
  ICML 2008, pp. 1176-1183.
• Learning predictive state representations using non-blind policies
  Michael Bowling, Peter McCracken, Michael James, James Neufeld, Dana Wilkinson
  ICML 2006, pp. 129-136.
• Predictive state representations with options
  Britton Wolfe, Satinder Singh
  ICML 2006, pp. 1025-1032.
• Learning predictive state representations in dynamical systems without reset
  Britton Wolfe, Michael R. James, Satinder Singh
  ICML 2005, pp. 980-987.
• Learning predictive representations from a history
  Eric Wiewiora
  ICML 2005, pp. 964-971.

動的環境 †

• Dealing with non-stationary environments using context detection
  Bruno C. da Silva, Eduardo W. Basso, Ana L. C. Bazzan, Paulo M. Engel
  ICML 2006, pp. 217-224.

価値関数近似 †

• Constructing basis functions from directed graphs for value function approximation
  Jeff Johns, Sridhar Mahadevan
  ICML 2007, pp. 385-392.
• Learning state-action basis functions for hierarchical MDPs
  Sarah Osentoski, Sridhar Mahadevan
  ICML 2007, pp. 705-712.
• Analyzing feature generation for value-function approximation
  Ronald Parr, Christopher Painter-Wakefield, Lihong Li, Michael Littman
  ICML 2007, pp. 737-744.
• Tracking value function dynamics to improve reinforcement learning with piecewise linear function approximation
  Chee Wee Phua, Robert Fitch
  ICML 2007, pp. 751-758.
• Automatic basis function construction for approximate dynamic programming and reinforcement learning
  Philipp W. Keller, Shie Mannor, Doina Precup
  ICML 2006, pp. 449-456.

連続的行動空間 †

• Binary action search for learning continuous-action control policies
  Jason Pazis and Michail Lagoudakis
  ICML 2009, pp. 793-800.

大規模状態空間 †

• Bayesian sparse sampling for on-line reward optimization
  Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans
  ICML 2005, pp. 956-963.
• Proto-value functions: developmental reinforcement learning
  Sridhar Mahadevan
  ICML 2005, pp. 553-560.

探査 †

• The adaptive k-meteorologists problem and its application to structure learning and feature selection in reinforcement learning
  Carlos Diuk, Lihong Li and Bethany Leffler
  ICML 2009, pp. 249-256.
• Near-Bayesian exploration in polynomial time
  J. Zico Kolter and Andrew Ng
  ICML 2009, pp. 513-520.
• Optimistic initialization and greediness lead to polynomial time learning in factored MDPs
  Istvan Szita and Andras Lorincz
  ICML 2009, pp. 1001-1008.
• Hoeffding and Bernstein races for selecting policies in evolutionary direct policy search
  Verena Heidrich-Meisner and Christian Igel
  ICML 2009, pp. 401-408.
• The many faces of optimism: a unifying approach
  Istvan Szita and Andras Lorincz
  ICML 2008, pp. 1048-1055.
• Reinforcement learning in the presence of rare events
  Jordan Frank, Shie Mannor, and Doina Precup
  ICML 2008, pp. 336-343.
• Percentile optimization in uncertain Markov decision processes with application to efficient exploration
  Erick Delage, Shie Mannor
  ICML 2007, pp. 225-232.
• An intrinsic reward mechanism for efficient exploration
  Özgür Şimşek, Andrew G. Barto
  ICML 2006, pp. 833-840.
• Qualitative reinforcement learning
  Arkady Epshteyn, Gerald DeJong
  ICML 2006, pp. 305-312.
• Experience-efficient learning in associative bandit problems
  Alexander L. Strehl, Chris Mesterharm, Michael L. Littman, Haym Hirsh
  ICML 2006, pp. 889-896.
• Exploration and apprenticeship learning in reinforcement learning
  Pieter Abbeel, Andrew Y. Ng
  ICML 2005, pp. 1-8.

行動規則評価 †

• A semiparametric statistical approach to model-free policy evaluation
  Tsuyoshi Ueno, Motoaki Kawanabe, Takeshi Mori, Shin-Ichi Maeda, and Shin Ishii
  ICML 2008, pp. 1072-1079.
• Exploration scavenging
  John Langford, Alexander Strehl, and Jennifer Wortman
  ICML 2008, pp. 528-535.
• Fast direct policy evaluation using multiscale analysis of Markov diffusion processes
  Mauro Maggioni, Sridhar Mahadevan
  ICML 2006, pp. 601-608.

学習分析 †

• A worst-case comparison between temporal difference and residual gradient with linear function approximation
  Lihong Li
  ICML 2008, pp. 560-567.
• An analysis of linear models, linear value-function approximation, and feature selection for reinforcement learning
  Ronald Parr, Lihong Li, Gavin Taylor, Christopher Painter-Wakefield
  ICML 2008, pp. 752-759.
• An analysis of reinforcement learning with function approximation
  Francisco Melo, Sean Meyn, and Isabel Ribeiro
  ICML 2008, pp. 664-671.
• A theoretical analysis of Model-Based Interval Estimation
  Alexander L. Strehl, Michael L. Littman
  ICML 2005, pp. 856-863.
• Relating reinforcement learning performance to classification performance
  John Langford, Bianca Zadrozny
  ICML 2005, pp. 473-480.

勾配法 †

• Fast gradient-descent methods for temporal-difference learning with linear function approximation
  Richard S. Sutton, Hamid Reza Maei, Doina Precup, Shalabh Bhatnagar, David Silver, Csaba Szepesvari, Eric Wiewiora
  ICML 2009, pp. 993-1000.
• Non-parametric policy gradients: a unified treatment of propositional and relational domains
  Kristian Kersting and Kurt Driessens
  ICML 2008, pp. 456-463.

TD学習 †

• Proto-predictive representation of states with simple recurrent temporal-difference networks
  Takaki Makino
  ICML 2009, pp. 697-704.
• Regularization and feature selection in least-squares temporal difference learning
  J. Zico Kolter and Andrew Ng
  ICML 2009, pp. 521-528.
• Kernelized value function approximation for reinforcement learning
  Gavin Taylor and Ronald Parr
  ICML 2009, pp. 1017-1024.
• Constraint relaxation in approximate linear programs
  Marek Petrik and Shlomo Zilberstein
  ICML 2009, pp. 809-816.
• Preconditioned temporal difference learning
  Hengshuai Yao and Zhi-Qiang Liu
  ICML 2008, pp. 1208-1215.
• PAC model-free reinforcement learning
  Alexander L. Strehl, Lihong Li, Eric Wiewiora, John Langford, Michael L. Littman
  ICML 2006, pp. 881-888.
• Reinforcement learning with Gaussian processes
  Yaakov Engel, Shie Mannor, Ron Meir
  ICML 2005, pp. 201-208.
• TD(λ) networks: temporal-difference networks with eligibility traces
  Brian Tanner, Richard S. Sutton
  ICML 2005, pp. 888-895.

アクター・クリティック †

• Bayesian actor-critic algorithms
  Mohammad Ghavamzadeh, Yaakov Engel
  ICML 2007, pp. 297-304.

モデル・ベースド †

• An analytic solution to discrete Bayesian reinforcement learning
  Pascal Poupart, Nikos Vlassis, Jesse Hoey, Kevin Regan
  ICML 2006, pp. 697-704.
• Using inaccurate models in reinforcement learning
  Pieter Abbeel, Morgan Quigley, Andrew Y. Ng
  ICML 2006, pp. 1-8.

その他・未分類 †

• Discovering options from example trajectories
  Peng Zang, Peng Zhou, David Minnen and Charles Isbell
  ICML 2009, pp. 1217-1224.
• An object-oriented representation for efficient reinforcement learning
  Carlos Diuk, Andre Cohen, and Michael Littman
  ICML 2008, pp. 240-247.
• Online kernel selection for Bayesian reinforcement learning
  Joseph Reisinger, Peter Stone, and Risto Miikkulainen
  ICML 2008, pp. 816-823.