DEALing with Image Reconstruction: Deep Attentive Least Squares
Abstract
State-of-the-art image reconstruction often relies on complex, abundantly parameterized deep architectures. We propose an alternative: a data-driven reconstruction method inspired by the classic Tikhonov regularization. Our approach iteratively refines intermediate reconstructions by solving a sequence of quadratic problems. These updates have two key components: (i) learned filters to extract salient image features; and (ii) an attention mechanism that locally adjusts the penalty of the filter responses. Our method matches leading plug-and-play and learned regularizer approaches in performance while offering interpretability, robustness, and convergent behavior. In effect, we bridge traditional regularization and deep learning with a principled reconstruction approach.
Cite
Text
Pourya et al. "DEALing with Image Reconstruction: Deep Attentive Least Squares." Proceedings of the 42nd International Conference on Machine Learning, 2025.Markdown
[Pourya et al. "DEALing with Image Reconstruction: Deep Attentive Least Squares." Proceedings of the 42nd International Conference on Machine Learning, 2025.](https://mlanthology.org/icml/2025/pourya2025icml-dealing/)BibTeX
@inproceedings{pourya2025icml-dealing,
title = {{DEALing with Image Reconstruction: Deep Attentive Least Squares}},
author = {Pourya, Mehrsa and Kobler, Erich and Unser, Michael and Neumayer, Sebastian},
booktitle = {Proceedings of the 42nd International Conference on Machine Learning},
year = {2025},
pages = {49689-49708},
volume = {267},
url = {https://mlanthology.org/icml/2025/pourya2025icml-dealing/}
}