Constrained GPI for Zero-Shot Transfer in Reinforcement Learning.
Jaekyeom KimSeohong ParkGunhee KimPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- transfer learning
- function approximation
- model free
- robotic control
- temporal difference
- state space
- learning algorithm
- markov decision processes
- information systems
- database
- dynamic load balancing
- robot control
- reinforcement learning algorithms
- action selection
- optimal control
- training data
- learning tasks
- dynamic programming
- learning process
- decision trees
- knowledge transfer
- generative model
- text classification
- neural network
- case study
- function approximators
- multi agent