Kurdyka- Lojasiewicz exponent via inf-projection

Peiran Yu The Hong Kong Polytechnic University, Guoyin Li University of New South Wales Tingkei Pong The Hong Kong Polytechnic University,

Numerical Analysis and Scientific Computing Numerical Linear Algebra Optimization and Control mathscidoc:2108.25006

Foundations of Computational Mathematics, 2021.7
Kurdyka–Łojasiewicz (KL) exponent plays an important role in estimating the convergence rate of many contemporary first-order methods. In particular, a KL exponent of 1/2 for a suitable potential function is related to local linear convergence. Nevertheless, KL exponent is in general extremely hard to estimate. In this paper, we show under mild assumptions that KL exponent is preserved via inf-projection. Inf-projection is a fundamental operation that is ubiquitous when reformulating optimization problems via the lift-and-project approach. By studying its operation on KL exponent, we show that the KL exponent is 1/2 for several important convex optimization models, including some semidefinite-programming-representable functions and some functions that involve C2-cone reducible structures, under conditions such as strict complementarity. Our results are applicable to concrete optimization models such as group-fused Lasso and overlapping group Lasso. In addition, for nonconvex models, we show that the KL exponent of many difference-of-convex functions can be derived from that of their natural majorant functions, and the KL exponent of the Bregman envelope of a function is the same as that of the function itself. Finally, we estimate the KL exponent of the sum of the least squares function and the indicator function of the set of matrices of rank at most k.
First-order methods · Convergence rate · Kurdyka–Łojasiewicz inequality · Kurdyka–Łojasiewicz exponent · Inf-projection
[ Download ] [ 2021-08-23 16:10:20 uploaded by gyli ] [ 157 downloads ] [ 0 comments ]
  • https://doi.org/10.1007/s10208-021-09528-6
  title={Kurdyka- Lojasiewicz exponent via inf-projection},
  author={Peiran Yu, Guoyin Li, and Tingkei Pong},
  booktitle={Foundations of Computational Mathematics},
Peiran Yu, Guoyin Li, and Tingkei Pong. Kurdyka- Lojasiewicz exponent via inf-projection. 2021. In Foundations of Computational Mathematics. http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20210823161020106125866.
Please log in for comment!
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved