Efficient Approaches for Escaping Higher Order Saddle Points in Non-Convex Optimization
Abstract
Local search heuristics for non-convex optimizations are popular in applied machine learning. However, in general it is hard to guarantee that such algorithms even converge to a local minimum, due to the existence of complicated saddle point structures in high dimensions. Many functions have degenerate saddle points such that the first and second order derivatives cannot distinguish them with local optima. In this paper we use higher order derivatives to escape these saddle points: we design the first efficient algorithm guaranteed to converge to a third order local optimum (while existing techniques are at most second order). We also show that it is NP-hard to extend this further to finding fourth order local optima.
Cite
Text
Anandkumar and Ge. "Efficient Approaches for Escaping Higher Order Saddle Points in Non-Convex Optimization." Annual Conference on Computational Learning Theory, 2016.Markdown
[Anandkumar and Ge. "Efficient Approaches for Escaping Higher Order Saddle Points in Non-Convex Optimization." Annual Conference on Computational Learning Theory, 2016.](https://mlanthology.org/colt/2016/anandkumar2016colt-efficient/)BibTeX
@inproceedings{anandkumar2016colt-efficient,
title = {{Efficient Approaches for Escaping Higher Order Saddle Points in Non-Convex Optimization}},
author = {Anandkumar, Animashree and Ge, Rong},
booktitle = {Annual Conference on Computational Learning Theory},
year = {2016},
pages = {81-102},
url = {https://mlanthology.org/colt/2016/anandkumar2016colt-efficient/}
}