HMRL: Hyper-Meta Learning for Sparse Reward Reinforcement Learning Problem.
Yun HuaXiangfeng WangBo JinWenhao LiJunchi YanXiaofeng HeHongyuan ZhaPublished in: KDD (2021)
Keyphrases
- meta learning
- reinforcement learning
- learning tasks
- inductive learning
- meta knowledge
- model selection
- machine learning algorithms
- machine learning
- decision trees
- transfer learning
- learning algorithm
- reward function
- high dimensional
- state space
- data mining
- markov decision processes
- feature selection
- optimal policy
- base classifiers
- multi agent
- learning process
- text classification
- supervised learning
- text categorization
- case study
- active learning
- xml documents
- expert systems