Meta-Reinforcement Learning in Broad and Non-Parametric Environments.
Zhenshan BingLukas KnakFabrice Oliver RobinKai HuangAlois C. KnollPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- machine learning
- function approximation
- dynamic environments
- state space
- real world
- optimal policy
- multi agent environments
- markov decision processes
- complex environments
- meta level
- model free
- website
- policy search
- artificial intelligence
- action selection
- temporal difference
- highly dynamic
- temporal difference learning
- semi supervised
- transfer learning
- model selection
- real time
- dynamic programming
- hidden markov models
- learning process
- multi agent
- case study
- web services
- search engine
- neural network