AISTATS 2021poster13 citations

On the Absence of Spurious Local Minima in Nonlinear Low-Rank Matrix Recovery Problems

Yingjie Bi, Javad Lavaei

Abstract

The restricted isometry property (RIP) is a well-known condition that guarantees the absence of spurious local minima in low-rank matrix recovery problems with linear measurements. In this paper, we introduce a novel property named bound difference property (BDP) to study low-rank matrix recovery problems with nonlinear measurements. Using RIP and BDP jointly, we propose a new criterion to certify the nonexistence of spurious local minima in the rank-1 case, and prove that it leads to a much stronger theoretical guarantee than the existing bounds on RIP.

BibTeX
@InProceedings{pmlr-v130-bi21a,
  title = 	 { On the Absence of Spurious Local Minima in Nonlinear Low-Rank Matrix Recovery Problems },
  author =       {Bi, Yingjie and Lavaei, Javad},
  booktitle = 	 {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {379--387},
  year = 	 {2021},
  editor = 	 {Banerjee, Arindam and Fukumizu, Kenji},
  volume = 	 {130},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {13--15 Apr},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v130/bi21a/bi21a.pdf},
  url = 	 {https://proceedings.mlr.press/v130/bi21a.html},
  abstract = 	 { The restricted isometry property (RIP) is a well-known condition that guarantees the absence of spurious local minima in low-rank matrix recovery problems with linear measurements. In this paper, we introduce a novel property named bound difference property (BDP) to study low-rank matrix recovery problems with nonlinear measurements. Using RIP and BDP jointly, we propose a new criterion to certify the nonexistence of spurious local minima in the rank-1 case, and prove that it leads to a much stronger theoretical guarantee than the existing bounds on RIP. }
}