Exchangeable Input Representations for Reinforcement Learning.
John MernDorsa SadighMykel J. KochenderferPublished in: ACC (2020)
Keyphrases
- reinforcement learning
- machine learning
- function approximation
- input data
- supervised learning
- markov decision processes
- multiple representations
- learning process
- hidden markov models
- state space
- website
- artificial intelligence
- transfer learning
- database
- learning classifier systems
- model free
- action selection
- reinforcement learning algorithms