Name
Abstract | ||
|---|---|---|
Parameter-specific adaptive learning rate methods are computationally efficient ways to reduce the ill-conditioning problems encountered when training large deep networks. Following recent work that strongly suggests that most of the critical points encountered when training such networks are saddle points, we find how considering the presence of negative eigenvalues of the Hessian could help us design better suited adaptive learning rate schemes. We show that the popular Jacobi preconditioner has undesirable behavior in the presence of both positive and negative curvature, and present theoretical and empirical evidence that the so-called equilibration preconditioner is comparatively better suited to non-convex problems. We introduce a novel adaptive learning rate scheme, called ESGD, based on the equilibration preconditioner. Our experiments show that ESGD performs as well or better than RMSProp in terms of convergence speed, always clearly improving over plain stochastic gradient descent. |
Year | Venue | Field |
|---|---|---|
2015 | Annual Conference on Neural Information Processing Systems | Convergence (routing),Mathematical optimization,Stochastic gradient descent,Saddle point,Preconditioner,Computer science,Hessian matrix,Critical point (mathematics),Adaptive learning,Eigenvalues and eigenvectors |
DocType | Volume | ISSN |
Conference | 28 | 1049-5258 |
Citations | PageRank | References |
52 | 2.49 | 9 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Dauphin, Yann N. | 1 | 979 | 49.26 |
| Harm de Vries | 2 | 239 | 12.50 |
| Yoshua Bengio | 3 | 42677 | 3039.83 |