Linearly convergent bilevel optimization with single-step inner methods

被引:0
作者
Ensio Suonperä
Tuomo Valkonen
机构
[1] University of Helsinki,Department of Mathematics and Statistics
[2] ModeMat,undefined
[3] Escuela Politécnica Nacional,undefined
来源
Computational Optimization and Applications | 2024年 / 87卷
关键词
Bilevel optimization; Nonsmooth; Inverse problems; Forward-backward;
D O I
暂无
中图分类号
学科分类号
摘要
We propose a new approach to solving bilevel optimization problems, intermediate between solving full-system optimality conditions with a Newton-type approach, and treating the inner problem as an implicit function. The overall idea is to solve the full-system optimality conditions, but to precondition them to alternate between taking steps of simple conventional methods for the inner problem, the adjoint equation, and the outer problem. While the inner objective has to be smooth, the outer objective may be nonsmooth subject to a prox-contractivity condition. We prove linear convergence of the approach for combinations of gradient descent and forward-backward splitting with exact and inexact solution of the adjoint equation. We demonstrate good performance on learning the regularization parameter for anisotropic total variation image denoising, and the convolution kernel for image deconvolution.
引用
收藏
页码:571 / 610
页数:39
相关论文
共 77 条
[1]  
Bard JF(1982)An explicit solution to the multi-level programming problem Comput. Oper. Res. 9 77-100
[2]  
Falk JE(2012)Solving bilevel programs with the KKT-approach Math. Program. 138 309-332
[3]  
Allende GB(2021)Gauss-newton-type methods for bilevel optimization Comput. Optim. Appl. 78 793-824
[4]  
Still G(2013)Application of particle swarm optimization based on CHKS smoothing function for solving nonlinear bilevel programming problem Appl. Math. Comput. 219 4332-4339
[5]  
Fliege J(1995)On bilevel programming, Part I: general nonlinear cases Math. Program. 70 47-72
[6]  
Tin A(2013)Image denoising: learning noise distribution via PDE-constrained optimization Inverse Prob. Imag. 7 1183-1214
[7]  
Zemkoho A(2013)A bilevel optimization approach for parameter learning in variational models SIAM J. Imag. Sci. 6 938-983
[8]  
Jiang Y(2018)A bilevel approach for parameter learning in inverse problems Inverse Prob. 34 515-4321
[9]  
Li X(2017)Optimal selection of the regularization function in a weighted total variation model. Part II: algorithm, its analysis and numerical tests J. Math. Imag. Vis. 59 4310-145
[10]  
Huang C(2020)Learning the sampling pattern for MRI IEEE Trans. Med. Imaging 39 120-149