ICLR 2026poster0 citations
Divergence-Free Neural Networks with Application to Image Denoising
Sébastien Herbreteau, Etienne Meunier
Abstract
We introduce a resource-efficient neural network architecture with zero divergence by design, adapted for high-dimensional problems. Our method is directly applicable to image denoising, for which divergence-free estimators are particularly well-suited for self-supervised learning, in accordance with Stein's unbiased risk estimation theory. Comparisons of our parameterization on popular denoising datasets demonstrate that it retains sufficient expressivity to remain competitive with other divergence-based approaches, while outperforming its counterparts when the noise level is unknown and varies across the training data.
image denoisingdivergenceStein's unbiased risk estimateself-supervised learningincompressible vector fields
BibTeX
@inproceedings{
herbreteau2026divergencefree,
title={Divergence-Free Neural Networks with Application to Image Denoising},
author={S{\'e}bastien Herbreteau and Etienne Meunier},
booktitle={The Fourteenth International Conference on Learning Representations},
year={2026},
url={https://openreview.net/forum?id=a5lL1ygtkG}
}