Toward Generalization of Automated Temporal Abstraction to Partially Observable Reinforcement Learning.
Erkin ÇildenFaruk PolatPublished in: IEEE Trans. Cybern. (2015)
Keyphrases
- partially observable
- reinforcement learning
- temporal abstractions
- state space
- partial observability
- markov decision processes
- partially observable environments
- partially observable domains
- markov decision problems
- reinforcement learning algorithms
- learning algorithm
- optimal policy
- model free
- reward function
- dynamical systems
- multi agent
- machine learning
- medical images
- optimal control
- dynamic programming
- function approximators
- video sequences
- decision making