Uncertainty Propagation for Efficient Exploration in Reinforcement Learning.
Alexander HansSteffen UdluftPublished in: ECAI (2010)
Keyphrases
- reinforcement learning
- function approximation
- partial observability
- sequential decision problems
- temporal difference
- machine learning
- learning algorithm
- uncertain data
- markov decision processes
- data sets
- wave propagation
- temporal difference learning
- learning problems
- optimal policy
- state space
- dynamic programming
- multi agent
- real time
- monte carlo
- supervised learning
- learning process
- case study
- decision making