Accelerating Reinforcement Learning with Learned Skill Priors.
Karl PertschYoungwoon LeeJoseph J. LimPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- learned from training data
- function approximation
- state space
- learned knowledge
- multi agent
- bayesian framework
- temporal difference
- optimal control
- reinforcement learning algorithms
- prior knowledge
- optimal policy
- markov decision processes
- data sets
- policy search
- action space
- prior information
- skills needed
- robotic control
- action selection
- learning problems
- maximum a posteriori
- transfer learning
- dynamic programming
- bayesian networks
- learning algorithm