CrossNorm: Normalization for Off-Policy TD Reinforcement Learning.
Aditya BhattMax ArgusArtemij AmiranashviliThomas BroxPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- temporal difference
- reinforcement learning algorithms
- temporal difference learning
- eligibility traces
- function approximation
- td learning
- state space
- learning algorithm
- markov decision processes
- optimal policy
- policy evaluation
- function approximators
- preprocessing
- evaluation function
- model free
- machine learning
- fixed point
- learning problems
- control problems
- real world
- policy iteration
- normalization method
- robotic control
- action selection
- transfer learning
- monte carlo
- supervised learning
- learning process
- neural network
- step size
- control policy
- reinforcement learning methods
- rl algorithms
- multi agent
- policy search
- decision making
- search engine
- temporal difference methods