Effects of Supervision, Population Size, and Self-Play on Multi-Agent Reinforcement Learning to Communicate.
Marina DubovaArseny MoskvichevPublished in: ALIFE (2020)
Keyphrases
- population size
- multi agent reinforcement learning
- genetic algorithm
- multi agent
- parameter settings
- multi agent learning
- mutation rate
- reinforcement learning
- markov chain
- stochastic games
- learning agents
- cooperative
- multi agent systems
- game playing
- learning process
- learning tasks
- convergence rate
- machine learning
- initial population
- learning agent