We live in a multi-agent world. To be successful in that world, intelligent agents need to learn to consider the agency of others. They will compete in marketplaces, cooperate in teams, communicate with others, coordinate their plans, and negotiate outcomes. Examples include self-driving cars interacting in traffic, personal assistants acting on behalf of humans and negotiating with other agents, swarms of unmanned aerial vehicles, financial trading systems, robotic teams, and household robots.

There has been great work on multi-agent learning in the past decade, but significant challenges remain, including the difficulty of learning an optimal model/policy from a partial signal, the exploration vs. exploitation dilemma, the scalability and effectiveness of learning algorithms, avoiding social dilemmas, learning emergent communication, learning to cooperate/compete in non-stationary environments with distributed simultaneously learning agents, and convergence guarantees.

Call for Papers

In addition to papers accepted to the 2020 session (now to be presented in the 2021 session due to the COVID-19 delay), we are pleased to announce an additional call for papers for 2021.

Authors can submit papers of 1-4 pages that will be reviewed by the organizing committee. We are looking for position papers that present a challenge or opportunity for MARL research, which should be on a topic the authors not only wish to interact on but also ‘work’ on with other participants during the symposium. We also welcome (preliminary) research papers that describe new perspectives to dealing with MARL challenges, but we are not looking for summaries of current research—papers should clearly state some limitation(s) of current methods and potential ways these could be overcome. Submissions will be handled through easychair.

We are interested in various challenges in multi-agent learning for this symposium, including:

• • Learning in sequential settings in dynamic environments (such as stochastic games, decentralized POMDPs and their variants)

  • Learning with partial observability

  • Exploration vs. exploitation techniques for multi-agent learning

  • Dynamics of multiple learners using (evolutionary) game theory

  • Learning with various communication limitations

  • Learning emergent communication

  • Learning in ad-hoc teamwork scenarios

  • Scalability through swarms vs. intelligent agents

  • Bayesian nonparametric methods for multi-agent learning

  • Deep learning and reinforcement learning methods for multi-agent learning

  • Transfer learning in multi-agent settings

  • Applications of multi-agent learning

  • Learning to cooperate/compete in non-stationary environments with distributed, simultaneously learning agents

  • Avoiding social dilemmas

The purpose of this symposium is to bring together researchers from machine learning, control, neuroscience, robotics, and multi-agent communities with the goal of broadening the scope of multi-agent learning research and addressing the fundamental issues that hinder the applicability of multi-agent learning for complex real world problems. This symposium will present a mix of invited sessions, contributed talks and a poster session with leading experts and active researchers from relevant fields. Furthermore, the symposium is designed to allow plenty of time for discussions and initiating collaborations.

Given the time lapsed since the 2020 session, we would like to offer authors who had their papers accepted for the March 2020 session the following options for the 2021 session:

1. Presenting their 2020 paper as-is in the new session.

2. Submitting a minor revision of their paper (i.e., minor updates/improvements, and no major change in topic). The organizing committee will subsequently verify the changes are minor (i.e., a minimal review).

3. Conduct a major revision of their 2020 paper. This will involve a full review by the PC.

4. Submitting a new paper altogether (and choosing to also present their 2020 paper as-is). This will involve a full review by the PC.

Submissions will be handled through easychair: https://easychair.org/conferences/?conf=sss21. The submission page will list the above four options under the 'Topics' header -- please ensure to select the appropriate option for your submission!

Please ensure your submissions are anonymous, with the main content a maximum of 4 pages (additional pages may be used for references and acknowledgements). Note that only PDF versions of the papers should be uploaded (no latex source necessary).

Author kit: https://www.aaai.org/Publications/Templates/AuthorKit21.zip

Important Dates

Submission: N̶o̶v̶e̶m̶b̶e̶r̶ ̶1̶s̶t̶,̶ ̶2̶0̶2̶0̶,̶ ̶2̶3̶:̶5̶9̶ ̶G̶M̶T̶ EXTENDED TO November 3rd, 2020, 23:59 GMT

• Notification: December 3rd, 2020

• Symposium: March 22-24, 2021

Challenge Descriptions

In the lead up to the workshop, we will contact both authors with accepted papers and others that indicated they will participate, asking them for a short description of the topic or challenge they would like to work on during the symposium. We will try to distill these into a number of core questions to work on. The ultimate goal of the symposium is to result in a number of joint position papers between participants that can grow into conference submissions within a year from the symposium.