Natural Emergence of Heterogeneous Strategies in Artificially Intelligent Competitive Teams
This work addresses the problem of strategy complexity in mixed cooperative-competitive environments for AI researchers, though it is incremental as it builds on existing methods like GNNs and RL.
The paper tackles the challenge of developing heterogeneous strategies in competitive multi-agent teams by introducing FortAttack, a competitive environment, and using Graph Neural Networks with Reinforcement Learning, which leads to the natural emergence of heterogeneous behaviors from homogeneous agents, enabling role replacement at test time.
Multi agent strategies in mixed cooperative-competitive environments can be hard to craft by hand because each agent needs to coordinate with its teammates while competing with its opponents. Learning based algorithms are appealing but many scenarios require heterogeneous agent behavior for the team's success and this increases the complexity of the learning algorithm. In this work, we develop a competitive multi agent environment called FortAttack in which two teams compete against each other. We corroborate that modeling agents with Graph Neural Networks and training them with Reinforcement Learning leads to the evolution of increasingly complex strategies for each team. We observe a natural emergence of heterogeneous behavior amongst homogeneous agents when such behavior can lead to the team's success. Such heterogeneous behavior from homogeneous agents is appealing because any agent can replace the role of another agent at test time. Finally, we propose ensemble training, in which we utilize the evolved opponent strategies to train a single policy for friendly agents.