Cooperative-Competitive Reinforcement Learning with History-Dependent Rewards.
Keyang HeBikramjit BanerjeePrashant DoshiPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- cooperative
- multi agent
- markov decision processes
- state space
- function approximation
- multi agent reinforcement learning
- multi agent systems
- reinforcement learning algorithms
- distributed problem solving
- model free
- learning problems
- game theory
- optimal policy
- temporal difference
- reward function
- reward shaping
- solve complex tasks
- learning process
- learning algorithm
- machine learning
- supervised learning
- least squares
- partially observable
- partially observable markov decision processes
- decision making
- cooperating agents
- robotic control
- data sets