Reinforcement Learning of Agent with a Staged View in Distance and Direction for the Pursuit Problem.
Tadayoshi YamamuraMotohide UmanoKazuhisa SetaPublished in: FUZZ-IEEE (2006)
Keyphrases
- reinforcement learning
- multi agent
- action selection
- multi agent systems
- multiagent systems
- learning agent
- reward function
- autonomous agents
- function approximation
- multi agent environments
- autonomous learning
- agent receives
- partially observable
- distance measure
- state abstraction
- action space
- learning algorithm
- learning capabilities
- exploration strategy
- dynamic environments
- model free
- software agents
- learning agents
- state action
- single agent
- state space
- reinforcement learning methods
- reinforcement learning agents
- reinforcement learning algorithms
- agent technology
- multiple agents
- mobile agents
- distance function
- distance metric
- learning process
- agent model
- machine learning
- robocup soccer
- dynamic programming
- multi agent reinforcement learning
- artificial agents
- intelligent agents
- markov decision processes
- temporal difference