Cooperative-Competitive Reinforcement Learning with History-Dependent Rewards.
Keyang HeBikramjit BanerjeePrashant DoshiPublished in: AAMAS (2021)
Keyphrases
- reinforcement learning
- cooperative
- multi agent
- function approximation
- markov decision processes
- multi agent systems
- state space
- reinforcement learning algorithms
- temporal difference
- reward shaping
- learning algorithm
- multi agent reinforcement learning
- optimal policy
- model free
- reward function
- neural network
- optimal control
- supervised learning
- hidden state
- partially observable
- cooperating agents
- policy search
- e learning
- distributed problem solving
- average reward
- function approximators
- policy iteration
- complex domains
- game theory
- learning process
- action selection
- cooperative learning
- learning problems