Learned Learning Rate Schedules for Deep Neural Network Training Using Reinforcement Learning.
Shreyas SubramanianVignesh GanapathiramanAly El GamalPublished in: Tiny Papers @ ICLR (2023)
Keyphrases
- learning rate
- neural network training
- training algorithm
- reinforcement learning
- learning algorithm
- convergence rate
- hidden layer
- neural network
- rapid convergence
- state space
- adaptive learning rate
- convergence speed
- optimization method
- markov decision processes
- particle swarm optimisation
- machine learning
- optimal policy
- control parameters
- training data
- dynamic programming
- feature vectors