Pith. sign in

REVIEW 3 cited by

First-order Methods Almost Always Avoid Saddle Points

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1710.07406 v1 pith:BTZLZMD7 submitted 2017-10-20 stat.ML cs.LGmath.OC

classification stat.MLcs.LGmath.OC
keywords avoiddescentfirst-ordermethodspointssaddlealmostaccess
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

We establish that first-order methods avoid saddle points for almost all initializations. Our results apply to a wide variety of first-order methods, including gradient descent, block coordinate descent, mirror descent and variants thereof. The connecting thread is that such algorithms can be studied from a dynamical systems perspective in which appropriate instantiations of the Stable Manifold Theorem allow for a global stability analysis. Thus, neither access to second-order derivative information nor randomness beyond initialization is necessary to provably avoid saddle points.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Extending the step-size restriction for gradient descent to avoid strict saddle points

    stat.ML 2019-08 conditional novelty 6.0 of 10

    Gradient descent almost surely avoids strict saddles for step sizes alpha < 2/L, provided the Hessian eigenvalue alpha^-1 occurs only on a measure-zero set.

  2. Deep Learning Theory Review: An Optimal Control and Dynamical Systems Perspective

    cs.LG 2019-08 conditional novelty 3.0 of 10

    A review that frames neural networks as dynamical systems, SGD as stochastic dynamics, and training as mean-field optimal control to unify deep learning theory.

  3. Distributed Stochastic Gradient Method for Non-Convex Problems with Applications in Supervised Learning

    math.OC 2019-08 conditional novelty 3.0 of 10

    A consensus-based distributed stochastic gradient algorithm is shown to converge in mean square to a critical point for non-convex problems, and it trains distributed neural networks on MNIST with accuracy comparable ...

Pith tools