Plan-based reward shaping for multi-agent reinforcement learning.
Sam DevlinDaniel KudenkoPublished in: Knowl. Eng. Rev. (2016)
Keyphrases
- multi agent reinforcement learning
- reward shaping
- reinforcement learning
- learning agent
- learning agents
- complex domains
- multi agent
- multi agent learning
- stochastic games
- reinforcement learning algorithms
- multi agent systems
- function approximation
- state space
- model free
- learning algorithm
- optimal control
- machine learning
- learning process
- cooperative
- learning tasks
- markov decision processes
- average reward
- transfer learning