# Learning inexact alternating minimization for convex problems with quadratic coupling

DevFeed: [Learning inexact alternating minimization for convex problems with quadratic coupling](<https://devfeed.tech/articles/learning-inexact-alternating-minimization-for-convex-problems-with-quadratic-coupling-125563.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.12544>)

Author: Paul H\\"ausner, Jevgenija Rudzusika, Jens Sj\\"olund, Ozan \\"Oktem

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [blocks](<https://devfeed.tech/tags/blocks.md>), [completion](<https://devfeed.tech/tags/completion.md>), [condition](<https://devfeed.tech/tags/condition.md>), [convex](<https://devfeed.tech/tags/convex.md>), [cost](<https://devfeed.tech/tags/cost.md>), [coupling](<https://devfeed.tech/tags/coupling.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [error](<https://devfeed.tech/tags/error.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [gap](<https://devfeed.tech/tags/gap.md>), [greedy](<https://devfeed.tech/tags/greedy.md>), [growth](<https://devfeed.tech/tags/growth.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [iteration](<https://devfeed.tech/tags/iteration.md>), [learning](<https://devfeed.tech/tags/learning.md>), [math-oc](<https://devfeed.tech/tags/math-oc.md>), [network](<https://devfeed.tech/tags/network.md>), [networks](<https://devfeed.tech/tags/networks.md>), [neural](<https://devfeed.tech/tags/neural.md>), [neural-networks](<https://devfeed.tech/tags/neural-networks.md>), [optimization](<https://devfeed.tech/tags/optimization.md>)

## AI overview

The paper learns an inexact alternating-minimization scheme by using neural networks to approximate each subproblem solution. For convex problems with quadratic coupling and a quadratic growth condition, it derives a worst-case optimality-gap bound that guides greedy per-iteration training, then combines that analysis with PAC-Bayes to bound expected gap over problem instances with high probability at a fixed horizon. On robust matrix completion and low-dose CT reconstruction with dictionary regularization, the method reports 5x-13x speedups over an inexact first-order baseline at matched accuracy; brief end-to-end fine-tuning raises the reported speedup to 30x and outperforms other learned schemes.

## Source excerpt

arXiv:2611.12544v1 Announce Type: cross Abstract: Alternating minimization is a standard tool for optimization problems with two coupled variable blocks. Its cost is dominated by the inner subproblem solvers, which rarely have a closed-form solution. In this paper, we propose to learn an inexact alternating scheme in which neural networks approximate the subproblem solution at each iteration. For convex problems with quadratic couplings between the variable blocks that satisfy a quadratic growth condition, we derive a worst-case bound on the optimality gap that exposes the per-iteration error that each network is trained to minimize in a greedy fashion. Combining the bound with a PAC-Bayes argument, we obtain a fixed-horizon, high-probability bound on the expected optimality gap over the distribution of problem instances. We apply the learned scheme to robust matrix completion and low-dose CT reconstruction with dictionary regularization where we achieve a 5x-13x speedup over the inexact first-order baseline at matched accuracy. A brief end-to-end fine-tuning stage extends this up to 30x and outperforms other learned schemes. The code is available at https://github.com/paulhausner/learned-alternating-minimization.