General Low-rank Matrix Optimization: Geometric Analysis and Sharper Bounds

被引:0
作者
Zhang, Haixiang [1 ]
Bi, Yingjie [2 ]
Lavaei, Javad [2 ]
机构
[1] Univ Calif Berkeley, Dept Math, Berkeley, CA 94704 USA
[2] Univ Calif Berkeley, Dept IEOR, Berkeley, CA 94704 USA
来源
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021) | 2021年 / 34卷
关键词
FACTORIZATION; ALGORITHMS; DESCENT;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper considers the global geometry of general low-rank minimization problems via the Burer-Monteiro factorization approach. For the rank-1 case, we prove that there is no spurious second-order critical point for both symmetric and asymmetric problems if the rank-2 RIP constant delta is less than 1/2. Combining with a counterexample with delta = 1/2, we show that the derived bound is the sharpest possible. For the arbitrary rank-r case, the same property is established when the rank-2r RIP constant delta is at most 1/3. We design a counterexample to show that the non-existence of spurious second-order critical points may not hold if delta is at least 1/2. In addition, for any problem with delta between 1/3 and 1/2, we prove that all second-order critical points have a positive correlation to the ground truth. Finally, the strict saddle property, which can lead to the polynomial-time global convergence of various algorithms, is established for both the symmetric and asymmetric problems when the rank-2r RIP constant delta is less than 1/3. The results of this paper significantly extend several existing bounds in the literature.
引用
收藏
页数:12
相关论文
共 36 条
  • [1] [Anonymous], 2018, NEURIPS
  • [2] [Anonymous], 2011, P 28 INT C MACH LEAR
  • [3] [Anonymous], 2016, C LEARNING THEORY
  • [4] [Anonymous], 2018, C LEARNING THEORY
  • [5] [Anonymous], 2016, ARXIV160500405
  • [6] Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods
    Attouch, Hedy
    Bolte, Jerome
    Svaiter, Benar Fux
    [J]. MATHEMATICAL PROGRAMMING, 2013, 137 (1-2) : 91 - 129
  • [7] Axiotis Kyriakos, 2020, P MACHINE LEARNING R, P452
  • [8] Bhojanapalli S, 2016, 30 C NEURAL INFORM P, V29
  • [9] Bi YJ, 2021, PR MACH LEARN RES, V130, P379
  • [10] NONCONVEX PHASE SYNCHRONIZATION
    Boumal, Nicolas
    [J]. SIAM JOURNAL ON OPTIMIZATION, 2016, 26 (04) : 2355 - 2377