Abstract | ||
---|---|---|
Stochastic gradient descent (SGD) still is the workhorse for many practical problems. However, it converges slow, and can be difficult to tune. It is possible to precondition SGD to accelerate its convergence remarkably. But many attempts in this direction either aim at solving specialized problems, or result in significantly more complicated methods than SGD. This paper proposes a new method to a... |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/TNNLS.2017.2672978 | IEEE Transactions on Neural Networks and Learning Systems |
Keywords | Field | DocType |
Eigenvalues and eigenfunctions,Optimization,Neural networks,Convergence,Newton method,Training,Acceleration | Gradient method,Convergence (routing),Mathematical optimization,Stochastic gradient descent,Preconditioner,Computer science,Recurrent neural network,Hessian matrix,Artificial intelligence,Artificial neural network,Machine learning,Newton's method | Journal |
Volume | Issue | ISSN |
29 | 5 | 2162-237X |
Citations | PageRank | References |
4 | 0.42 | 5 |
Authors | ||
1 |