Reinforcement learning supercharges redox flow batteries.
Yang CaoCher Tian SerMarta SkretaKjell JornerNathanael KusandaAlán Aspuru-GuzikPublished in: Nat. Mach. Intell. (2022)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- temporal difference
- learning algorithm
- markov decision processes
- optimal policy
- model free
- state space
- dynamic programming
- energy consumption
- learning process
- flow field
- information flow
- machine learning
- real time
- policy search
- action space
- temporal difference learning
- flow patterns
- energy efficient
- learning problems
- multi agent
- real world
- neural network