How to join

We have now shared links to the event/calendar with symposium registrants (if you have registered but have not received these links, please reach out to the symposium organizers and sss21@aaai.org ASAP). For convenience, we have set up Google Meets with identical links for both days (i.e., you can join the same meet and stay within it throughout each day to watch all talks).

Program

Please be mindful of timezones in the below program, which take into account daylight savings having begun in the US on March 14th, and are not starting in Europe until March 28th.

Day 1 (March 22, 2021)

Session 1 (Invited Talk)| 2-3 pm CET / 9-10am EDT

• • Georgios Piliouras: Beyond equilibrium: A dynamical systems approach to learning in games

Session 2 (Paper Presentations) | 3-4pm CET / 10-11am EDT

• • [15 minutes] William A. Dawson, Ruben Glatt, Edward Rusu, Braden C. Soper and Ryan A. Goldhahn, "Hybrid Information-driven Multi-agent Reinforcement Learning"

  • [15 minutes] Chirag Chhablani and Ian Kash, "Position Paper: Regret Minimization for Stateful, Cooperative Settings"

  • [15 minutes] Wolfram Barfuss, "Towards a unified treatment of the dynamics of collective learning"

  • [15 minutes] Xiaobai Ma, Jayesh K. Gupta and Mykel J. Kochenderfer, "Policy Representation in Continuous Action Games"

Session 3 (Paper Presentations) | 4-5pm CET / 11am-12pm EDT

• • [15 minutes] Mahak Goindani and Jennifer Neville, "Social Reinforcement Learning"

  • 45 minute Break

Session 4 (Invited Talk) | 5-6pm CET / 12pm-1pm EDT

• • Edward Lockhart: OpenSpiel Tutorial

Session 5 (Paper Presentations) | 6-7pm CET / 1-2pm EDT

• • [15 minutes] Clément Moulin-Frier and Pierre-Yves Oudeyer, "Multi-Agent Reinforcement Learning as a Computational Tool for Language Evolution Research: Historical Context and Future Challenges"

  • [15 minutes] Paul Van Eecke and Katrien Beuls, "Re-conceptualising the Language Game Paradigm in the Framework of Multi-Agent Reinforcement Learning"

  • [15 minutes] Kyle Tilbury and Jesse Hoey, "Multi-Agent Reinforcement Learning and Human Social Factors in Climate Change Mitigation"

  • [15 minutes] Elad Liebman, Alexandre Ardel, Shashank Bassi, Jacob Riedel and Edgars Vitolins, "Autonomous Multiagent Aviation: Challenges and Opportunities"

Day 2 (March 23, 2021)

Session 6 (Invited Talk) | 2-3 pm CET / 9-10am EDT

• • Romuald Elie: Mean Field Learning with Infinitely Many Agents

Session 7 (Paper Presentations) | 3-4pm CET / 10-11am EDT

• • [15 minutes] Bengisu Güresti and Nazim Kemal Ure, "Evaluating Generalization and Transfer Capacity of Multi-Agent ReinforcementLearning Across Variable Number of Agents"

  • [15 minutes] Dong-Ki Kim, Miao Liu, Matthew Riemer, Golnaz Habibi, Sebastian Lopez-Cot, Samir Wadhwania, Gerald Tesauro and Jonathan How, "A Policy Gradient Theorem for Learning to Learn in Multiagent Reinforcement Learning"

  • [15 minutes] Tushant Jha, "The Role of Artificial Institutions in Multi Agent Learning: A Research Agenda"

  • [15 minutes] William Birmingham, Sarah Dumnich and Britton Wolfe, "Separate worlds, separate clocks: issues in asynchronous MDPs"

Session 8 (Paper Presentations) | 4-5pm CET / 11am-12pm EDT

• • [15 minutes] Theocharis Kravaris and George Vouros, "Scalable Deep Multi-Agent Reinforcement Learning"

  • 45 minute Break

Session 9 (Invited Talk) | 5-6pm CET / 12-1pm EDT

• • Thore Graepel: From AlphaGo to MuZero - Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model

Session 10 (Paper Presentations) | 6-7pm CET / 1-2pm EDT

• • [15 minutes] Stephen Cranefield, "Learning Norms in Multi-Agent Systems: A Challenge to the MARL Community"

  • [15 minutes] Alexander Shmakov, John Lanier, Stephen McAleer, Rohan Achar, Christina Lopes and Pierre Baldi, "ColosseumRL: A Framework for Multiagent Reinforcement Learning in N-Player Games"

  • [15 minutes] Stefan Heidekrüger, Nils Kohring, Paul Sutterer and Martin Bichler, "Multiagent Learning for Equilibrium Computation in Auction Markets"

  • [15 minutes] Qi Zhang, "Meta-Learning Multi-Agent Communication"

Session 11 | 7-7:30 pm CET / 2-2:30pm EDT

• • Closing remarks from the Organizers