Learning Macro-actions for State-Space Planning.
Sandra Castellanos-PaezDamien PellierHumbert FiorinoSylvie PestyPublished in: CoRR (2016)
Keyphrases
- macro actions
- state space
- reinforcement learning
- markov decision processes
- temporally extended
- stochastic domains
- learning algorithm
- predictive model
- planning domains
- learning process
- learned knowledge
- dynamic programming
- planning problems
- heuristic search
- markov chain
- markov decision problems
- domain independent
- neural network
- state variables
- plan recognition
- reinforcement learning algorithms
- multi agent
- search space