Importance Weighted Transfer of Samples in Reinforcement Learning.
Andrea TirinzoniAndrea SessaMatteo PirottaMarcello RestelliPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- transfer learning
- markov decision processes
- training samples
- real time
- error minimization
- weighted sum
- function approximation
- state space
- robotic control
- database
- previously learned
- sample points
- model free
- relative importance
- data sets
- data mining
- learning algorithm
- multi agent
- least squares
- optimal control
- knowledge transfer
- reinforcement learning algorithms
- training set
- supervised learning
- learning process