Ray Interference: a Source of Plateaus in Deep Reinforcement Learning.
Tom SchaulDiana BorsaJoseph ModayilRazvan PascanuPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- state space
- multi agent
- markov decision processes
- dynamic programming
- optimal policy
- robotic control
- action selection
- multiple sources
- transfer learning
- learning algorithm
- supervised learning
- semi supervised
- mobile robot
- learning process
- artificial neural networks
- optimal control
- search algorithm
- hill climbing
- reinforcement learning algorithms
- learning agent
- reinforcement learning methods
- data sets