DEALing with Image Reconstruction: Deep Attentive Least Squares
M. Pourya, E. Kobler, M. Unser, S. Neumayer
Proceedings of the Forty-Second International Conference on Machine Learning (ICML'25), Vancouver BC, Canada, July 13-19, 2025, in press.
Please do not bookmark the In Press papers as content and presentation may differ from the published version.
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.