: Deep Case-based Policy Inference for knowledge transfer in Reinforcement Learning.
Ruben GlattFelipe Leno da SilvaReinaldo Augusto da Costa BianchiAnna Helena Reali CostaPublished in: Expert Syst. Appl. (2020)
Keyphrases
- knowledge transfer
- reinforcement learning
- transfer learning
- optimal policy
- policy search
- transferring knowledge
- markov decision process
- knowledge sharing
- action selection
- learning tasks
- state space
- control policy
- previously learned
- action space
- function approximation
- cross domain
- reward function
- function approximators
- policy gradient
- learning algorithm
- markov decision processes
- mutual adaptation
- reinforcement learning algorithms
- model free
- transfer knowledge
- data sets
- social interaction
- nearest neighbor
- active learning
- training data
- similarity measure