Continual Reinforcement Learning deployed in Real-life using Policy Distillation and Sim2Real Transfer.
René TraoréHugo Caselles-DupréTimothée LesortTe SunNatalia Díaz RodríguezDavid FilliatPublished in: CoRR (2019)
Keyphrases
- real life
- reinforcement learning
- optimal policy
- transfer learning
- policy search
- action selection
- real world
- application domains
- state space
- markov decision process
- markov decision processes
- function approximation
- neural network
- control policy
- state action
- agent receives
- actor critic
- policy gradient
- markov decision problems
- control policies
- action space
- policy iteration
- reinforcement learning algorithms
- machine learning
- learning algorithm