Agent Societies
Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games Fanqi Kong 2, 1 Yizhe Huang
Real-world multi-agent scenarios often involve mixed motives, demanding altruistic agents capable of self-protection against potential exploitation.
Multi-Agent Coordination via Multi-Level Communication
SeqComm are guaranteed to improve monotonically and converge.
N-Agent Ad Hoc Teamwork
Current approaches to learning cooperative multi-agent behaviors assume relatively restrictive settings.
Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team Games
Team-Nash equilibrium (TNE) predicts the outcomes of such coordinated interactions.
Counterfactual Conservative Q Learning for Offline Multi-agent Reinforcement Learning Jianzhun Shao, Y un Qu
MARL in real scenarios is still challenging due to the same safety and efficiency concerns in single-agent setting, then it is worth conducting investigation for offline RL in multi-agent setting.
Scenario Diffusion: Controllable Driving Scenario Generation With Diffusion
Work done during an internship at Zoox 37th Conference on Neural Information Processing Systems (NeurIPS 2023).
Selectively Sharing Experiences Improves Multi-Agent Reinforcement Learning
We present a novel multi-agent RL approach, Selective Multi-Agent Prioritized Experience Relay, in which agents share with other agents a limited number of transitions they observe during training.