Continual Reinforcement Learning in 3D Non-stationary Environments.
Vincenzo LomonacoKaran DesaiEugenio CulurcielloDavide MaltoniPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- state space
- robotic control
- class imbalance
- reinforcement learning algorithms
- temporal difference
- learning process
- supervised learning
- markov decision processes
- machine learning
- robot control
- non stationary
- artificial intelligence
- learning algorithm
- transfer learning
- learning problems
- optimal control
- dynamic environments
- multi agent
- partially observable
- control problems
- reinforcement learning methods
- neural network
- direct policy search