Please use this identifier to cite or link to this item:
Title: Kurdyka-Lojasiewicz exponent via inf-projection
Authors: Yu, P
Li, G
Pong, TK 
Issue Date: 2021
Source: Foundations of computational mathematics, 2021,
Abstract: 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 12 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 12 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.
Keywords: First-order methods
Convergence rate
Kurdyka-Lojasiewicz inequality
Kurdyka-Lojasiewicz exponent
Publisher: Springer
Journal: Foundations of computational mathematics 
ISSN: 1615-3375
DOI: 10.1007/s10208-021-09528-6
Appears in Collections:Journal/Magazine Article

Open Access Information
Status embargoed access
Embargo End Date 2022-07-09
View full-text via PolyU eLinks SFX Query
Show full item record

Page views

Citations as of May 15, 2022


Citations as of May 12, 2022


Citations as of May 19, 2022

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.