SMiRL: Surprise Minimizing Reinforcement Learning in Unstable Environments.
Glen BersethDaniel GengColine Manon DevinNicholas RhinehartChelsea FinnDinesh JayaramanSergey LevinePublished in: ICLR (2021)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- model free
- state space
- dynamic environments
- markov decision processes
- multi agent environments
- highly dynamic
- robotic systems
- real world
- optimal policy
- computing environments
- dynamic programming
- learning process
- learning algorithm
- data mining
- temporal difference learning
- autonomous learning
- databases