Offline Meta-Reinforcement Learning with Advantage Weighting.
Eric MitchellRafael RafailovXue Bin PengSergey LevineChelsea FinnPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- real time
- machine learning
- multi agent
- optimal policy
- robotic control
- temporal difference learning
- reinforcement learning algorithms
- meta level
- markov decision processes
- mobile robot
- dynamic programming
- optimal control
- tf idf
- similarity measure
- model free
- action selection
- artificial intelligence
- robot control
- learning agents
- learning algorithm
- reinforcement learning methods
- databases