Target-Network Update Linked with Learning Rate Decay Based on Mutual Information and Reward in Deep Reinforcement Learning.
Chayoung KimPublished in: Symmetry (2023)
Keyphrases
- learning rate
- reinforcement learning
- mutual information
- learning algorithm
- convergence rate
- similarity measure
- image registration
- adaptive learning rate
- error function
- function approximation
- model free
- machine learning
- reinforcement learning algorithms
- power law
- convergence speed
- multilayer neural networks
- activation function
- convergence theorem
- bp neural network algorithm
- eligibility traces
- reinforcement learning methods
- state action
- temporal difference
- network architecture
- optimal policy
- genetic algorithm