You are not logged in |
Gentle introduction to multi-agent reinforcement learning
- Speaker(s)
- Michał Zawalski
- Affiliation
- Uniwerystet Warszawski
- Date
- March 3, 2022, 12:15 p.m.
- Information about the event
- meet.google.com/ooi-zxye-dxa
- Seminar
- Seminarium "Machine Learning"
Recent studies show some impressive applications of reinforcement learning algorithms in sequential decision-making problems. In my talk, I will focus on problems that involve controlling a group of agents, i.e. multi-agent reinforcement learning. Though that domain shares clear similarities with the standard, single-agent case, there are some specific issues.
I will start by introducing the reinforcement learning framework, the key concepts and simple approaches. Then I will describe some multi-agent tasks and highlight the difficulties that make standard RL algorithms fail. Finally, I will share some insights discovered in our work "Off-Policy Correction For Multi-Agent Reinforcement Learning", accepted for publication at AAMAS 2022.
Seminar will be held remotely.
