Sim2Real Transfer for Reinforcement Learning without Dynamics Randomization.
Manuel KasparJuan David Muñoz OsorioJürgen BockPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- transfer learning
- state space
- function approximation
- learning algorithm
- real life
- privacy preserving
- dynamic model
- databases
- information systems
- knowledge transfer
- stochastic approximation
- model free
- learning problems
- dynamical systems
- monte carlo
- optimal policy
- search space
- artificial intelligence
- machine learning