On the Stability of Gradient Descent for Large Learning Rate.
Alexandru CraciunDebarghya GhoshdastidarPublished in: CoRR (2024)
Keyphrases
- learning rate
- error function
- uniform convergence
- update rule
- natural gradient
- learning algorithm
- convergence rate
- hidden layer
- convergence speed
- multilayer neural networks
- adaptive learning rate
- loss function
- training algorithm
- cost function
- rapid convergence
- delta bar delta
- weight vector
- conjugate gradient
- objective function
- learning rules
- bp neural network algorithm
- feedforward neural networks
- neural network
- linear programming
- data mining