Dynamically writing coupled memories using a reinforcement learning agent, meeting physical bounds.
Théo JulesLaura MichelAdele DouinFrédéric LechenaultPublished in: CoRR (2022)
Keyphrases
- learning agent
- reinforcement learning
- state space
- reinforcement learning algorithms
- solving problems
- lower bound
- learning algorithm
- learning capabilities
- learning tasks
- function approximation
- selective perception
- learning process
- active learning
- reward function
- multi agent systems
- input output
- model free
- cooperative
- mixed initiative
- knowledge base
- neural network