pith. sign in

arxiv: 1902.02366 · v1 · pith:KNEIYOGEnew · submitted 2019-02-06 · 💻 cs.LG · math.OC· stat.ML

Negative eigenvalues of the Hessian in deep neural networks

classification 💻 cs.LG math.OCstat.ML
keywords deepnetworkseigenvalueshessianlossnegativeactiveappropriately
0
0 comments X
read the original abstract

The loss function of deep networks is known to be non-convex but the precise nature of this nonconvexity is still an active area of research. In this work, we study the loss landscape of deep networks through the eigendecompositions of their Hessian matrix. In particular, we examine how important the negative eigenvalues are and the benefits one can observe in handling them appropriately.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.