Influence Graph based Task Decomposition and State Abstraction in Reinforcement Learning.
Lasheng YuFei HongPengRen WangYang XuYong LiuPublished in: ICYCS (2008)
Keyphrases
- state abstraction
- reinforcement learning
- state space
- reinforcement learning agents
- hierarchical reinforcement learning
- markov decision processes
- path finding
- reinforcement learning algorithms
- function approximation
- semi supervised
- learning algorithm
- search space
- multi agent
- admissible heuristics
- initial state
- optimal control
- function approximators
- heuristic search
- monte carlo
- optimal policy
- markov chain