We are having the multi-agent reinforcement learning (MARL) reading group meeting starting from August 25, 2017. It would be a weekly discussion on Fridays (usually 12pm-1pm, subject to change). If you are interested in joining the reading group, please email your name, title, your home department at CMU as well as your research interests to aiandsocialgood@gmail.com and wait for approval. We expect everyone reads the paper before coming to the reading group and prepares a few questions for discussion.

Schedule:

• 12pm-1pm ET, November 17, 2017. Wean 4220.
  A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning
  Marc Lanctot, Vinicius Zambaldi, Audrunas Gruslys, Angeliki Lazaridou, Karl Tuyls, Julien Perolat, David Silver, Thore Graepel
• 12pm-1pm ET, November 10, 2017. Wean 4220.
  DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker
  Matej Moravčík, Martin Schmid, Neil Burch, Viliam Lisý, Dustin Morrill, Nolan Bard, Trevor Davis, Kevin Waugh, Michael Johanson, Michael Bowling
• 12pm-1pm ET, October 27, 2017. Wean 4220.
  Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments
  Maruan Al-Shedivat, Trapit Bansal, Yuri Burda, Ilya Sutskever, Igor Mordatch, Pieter Abbeel
• 12pm-1pm ET, October 20, 2017. Wean 4220.
  Fully Decentralized Policies for Multi-Agent Systems: An Information Theoretic Approach
  Roel Dobbe, David Fridovich-Keil, Claire Tomlin
• 12pm-1pm ET, October 13, 2017. Wean 4220.
  Plan-based reward shaping for multi-agent reinforcement learning
  SAM DEVLIN and DANIEL KUDENKO
• 12pm-1pm ET, October 6, 2017. Wean 4220.
  Towards Deep Learning Models Resistant to Adversarial Attacks
  Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, Adrian Vladu
• 12pm-1pm ET, September 29, 2017. Wean 4220.
  Deep Decentralized Multi-task Multi-Agent Reinforcement Learning under Partial Observability, ICML 2017
  Shayegan Omidshafiei, Jason Pazis, Christopher Amato, Jonathan P. How, John Vian

• 12pm-1pm ET, September 22, 2017. Wean 4220.
  Learning with Opponent-Learning Awareness
  Jakob N. Foerster, Richard Y. Chen, Maruan Al-Shedivat, Shimon Whiteson, Pieter Abbeel, Igor Mordatch
• 12pm-1pm ET, September 15, 2017. Wean 4220.
  Deep reinforcement learning from self-play in imperfect-information games
  Heinrich, Johannes, and David Silver. arXiv, 2016.
  Multiagent Bidirectionally-Coordinated Nets for Learning to Play StarCraft Combat Games
  Peng P, Yuan Q, Wen Y, et al. arXiv, 2017.
• 12pm-1pm ET, September 8, 2017. Wean 4220.
  Deep Recurrent Q-Learning for Partially Observable MDPs
  Matthew Hausknecht, Peter Stone
• 12pm-1pm ET, September 1, 2017. Wean 4126.
  A comprehensive survey of multi-agent reinforcement learning
  Busoniu L, Babuska R, De Schutter B. IEEE Transactions on Systems Man and Cybernetics Part C Applications and Reviews, 2008
• 1pm-2pm ET, August 25, 2017. Wean 4126.
  Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
  Ryan Lowe, Yi Wu, Aviv Tamar, Jean Harb, Pieter Abbeel, Igor Mordatch
  Animation (video):

Useful resources:

• Paper Collection of Multi-Agent Reinforcement Learning (MARL)