Reinforcement Learning with Partially Known World Dynamics
Christian R. SheltonPublished in: CoRR (2013)
Keyphrases
- reinforcement learning
- machine learning
- function approximation
- dynamic model
- markov decision processes
- dynamical systems
- learning problems
- active learning
- multi agent reinforcement learning
- highly nonlinear
- reinforcement learning algorithms
- learning process
- real world
- temporal difference
- artificial intelligence
- agent learns
- reinforcement learning methods
- world model
- optimal control
- decision making
- multi agent
- markov chain
- supervised learning
- state space
- evolutionary algorithm