State Space Construction for Behavior Acquisition in Multi Agent Environments with Vision and Action.
Eiji UchibeMinoru AsadaKoh HosodaPublished in: ICCV (1998)
Keyphrases
- multi agent environments
- state space
- reinforcement learning
- reinforcement learning algorithms
- multi agent
- autonomous agents
- multi agent systems
- action space
- markov decision processes
- heuristic search
- function approximation
- optimal policy
- state transition
- reinforcement learning methods
- dynamical systems
- initial state
- state variables
- model free
- reinforcement learning agents
- control problems
- markov chain
- reward signal
- temporal difference
- action selection
- particle filter
- search algorithm
- real valued
- virtual environment
- dynamic programming
- machine learning