Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation.
Arnulf JentzenTimo WeltiPublished in: Appl. Math. Comput. (2023)
Keyphrases
- stochastic gradient descent
- error analysis
- least squares
- neural network
- early stopping
- training speed
- matrix factorization
- loss function
- step size
- random forests
- back propagation
- error correction
- training algorithm
- regularization parameter
- training process
- support vector machine
- collaborative filtering
- image restoration
- training samples
- weight vector
- optical flow
- support vector
- cost function
- multiple kernel learning
- online algorithms
- pairwise