Reward shaping for knowledge-based multi-objective multi-agent reinforcement learning.
Patrick MannionSam DevlinJim DugganEnda HowleyPublished in: Knowl. Eng. Rev. (2018)
Keyphrases
- multi agent reinforcement learning
- multi objective
- reward shaping
- reinforcement learning
- learning agent
- evolutionary algorithm
- complex domains
- multi agent learning
- objective function
- function approximation
- learning agents
- multi agent
- reinforcement learning algorithms
- expert systems
- multi agent systems
- genetic algorithm
- state space
- learning process
- learning algorithm
- markov decision processes
- neural network
- stochastic games
- model free
- cooperative
- machine learning
- learning tasks
- supervised learning
- partially observable
- artificial intelligence