Learning cooperative behaviours in adversarial multi-agent systems

02/10/2023

∙

This work extends an existing virtual multi-agent platform called RoboSumo to create TripleSumo – a platform for investigating multi-agent cooperative behaviors in continuous action spaces, with physical contact in an adversarial environment. In this paper we investigate a scenario in which two agents, namely `Bug' and `Ant', must team up and push another agent `Spider' out of the arena. To tackle this goal, the newly added agent `Bug' is trained during an ongoing match between `Ant' and `Spider'. `Bug' must develop awareness of the other agents' actions, infer the strategy of both sides, and eventually learn an action policy to cooperate. The reinforcement learning algorithm Deep Deterministic Policy Gradient (DDPG) is implemented with a hybrid reward structure combining dense and sparse rewards. The cooperative behavior is quantitatively evaluated by the mean probability of winning the match and mean number of steps needed to win.

READ FULL TEXT

Learning cooperative behaviours in adversarial multi-agent systems

Multi-Agent Reinforcement Learning for Problems with Combined Individual and Team Reward

On the Robustness of Cooperative Multi-Agent Reinforcement Learning

Consolidation via Policy Information Regularization in Deep RL for Multi-Agent Games

On the Approximation of Cooperative Heterogeneous Multi-Agent Reinforcement Learning (MARL) using Mean Field Control (MFC)

Two-stage training algorithm for AI robot soccer

Evaluating Robustness of Cooperative MARL: A Model-based Approach

Learning to Simulate Self-Driven Particles System with Coordinated Policy Optimization

Learning cooperative behaviours in adversarial multi-agent systems

Related Research

Multi-Agent Reinforcement Learning for Problems with Combined Individual and Team Reward

On the Robustness of Cooperative Multi-Agent Reinforcement Learning

Consolidation via Policy Information Regularization in Deep RL for Multi-Agent Games

On the Approximation of Cooperative Heterogeneous Multi-Agent Reinforcement Learning (MARL) using Mean Field Control (MFC)

Two-stage training algorithm for AI robot soccer

Evaluating Robustness of Cooperative MARL: A Model-based Approach

Learning to Simulate Self-Driven Particles System with Coordinated Policy Optimization