Robust Task Representations for Offline Meta-Reinforcement Learning via Contrastive Learning.
Haoqi YuanZongqing LuPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- learning problems
- reinforcement learning methods
- learning analytics
- supervised learning
- learning tasks
- autonomous learning
- inductive inference
- knowledge acquisition
- learning experience
- background knowledge
- temporal difference
- automatically discovering
- learning capabilities
- function approximators
- learned knowledge
- data sets
- model free
- function approximation
- unsupervised learning
- online learning
- multi agent
- decision trees
- machine learning