MURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning.
Kevin LiAbhishek GuptaAshwin ReddyVitchyr H. PongAurick ZhouJustin YuSergey LevinePublished in: ICML (2021)
Keyphrases
- meta learning
- reinforcement learning
- learning tasks
- meta knowledge
- inductive learning
- machine learning
- model selection
- markov decision processes
- learning algorithm
- transfer learning
- machine learning algorithms
- state space
- decision trees
- feature selection
- multi agent
- reward function
- supervised learning
- artificial intelligence
- metamodel
- optimal policy
- multi class
- learning process
- base classifiers
- learning experience
- data sets
- inductive logic programming
- database systems
- data mining
- databases