English

Inexact Derivative-Free Optimization for Bilevel Learning

Optimization and Control 2020-12-10 v2 Computer Vision and Pattern Recognition Machine Learning Numerical Analysis Numerical Analysis Machine Learning

Abstract

Variational regularization techniques are dominant in the field of mathematical imaging. A drawback of these techniques is that they are dependent on a number of parameters which have to be set by the user. A by now common strategy to resolve this issue is to learn these parameters from data. While mathematically appealing this strategy leads to a nested optimization problem (known as bilevel optimization) which is computationally very difficult to handle. It is common when solving the upper-level problem to assume access to exact solutions of the lower-level problem, which is practically infeasible. In this work we propose to solve these problems using inexact derivative-free optimization algorithms which never require exact lower-level problem solutions, but instead assume access to approximate solutions with controllable accuracy, which is achievable in practice. We prove global convergence and a worstcase complexity bound for our approach. We test our proposed framework on ROFdenoising and learning MRI sampling patterns. Dynamically adjusting the lower-level accuracy yields learned parameters with similar reconstruction quality as highaccuracy evaluations but with dramatic reductions in computational work (up to 100 times faster in some cases).

Keywords

Cite

@article{arxiv.2006.12674,
  title  = {Inexact Derivative-Free Optimization for Bilevel Learning},
  author = {Matthias J. Ehrhardt and Lindon Roberts},
  journal= {arXiv preprint arXiv:2006.12674},
  year   = {2020}
}

Comments

Accepted to Journal of Mathematical Imaging and Vision

R2 v1 2026-06-23T16:32:26.438Z