Convergence of stochastic gradient descent under a local Lajasiewicz condition for deep neural networks.
Jing AnJianfeng LuPublished in: CoRR (2023)
Keyphrases
- stochastic gradient descent
- neural network
- step size
- least squares
- loss function
- convergence rate
- matrix factorization
- convergence speed
- random forests
- support vector machine
- online algorithms
- importance sampling
- regularization parameter
- weight vector
- support vector
- multiple kernel learning
- knn
- image restoration
- feature space
- training data
- decision trees