Recurrent neural network training with preconditioned stochastic gradient descent.
Xi-Lin LiPublished in: CoRR (2016)
Keyphrases
- stochastic gradient descent
- recurrent neural networks
- early stopping
- recurrent networks
- least squares
- matrix factorization
- step size
- loss function
- echo state networks
- neural network
- random forests
- feed forward
- support vector machine
- neural network structure
- hidden layer
- training algorithm
- weight vector
- importance sampling
- online algorithms
- artificial neural networks
- multiple kernel learning
- convergence speed
- regularization parameter
- pairwise