Robust Classification by Coupling Data Mollification with Label Smoothing

Markus Heinonen, Ba-Hien Tran, Michael Kampffmeyer, Maurizio Filippone
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:4960-4968, 2025.

Abstract

Introducing training-time augmentations is a key technique to enhance generalization and prepare deep neural networks against test-time corruptions. Inspired by the success of generative diffusion models, we propose a novel approach of coupling data mollification, in the form of image noising and blurring, with label smoothing to align predicted label confidences with image degradation. The method is simple to implement, introduces negligible overheads, and can be combined with existing augmentations. We demonstrate improved robustness and uncertainty quantification on the corrupted image benchmarks of CIFAR, TinyImageNet and ImageNet datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-heinonen25a, title = {Robust Classification by Coupling Data Mollification with Label Smoothing}, author = {Heinonen, Markus and Tran, Ba-Hien and Kampffmeyer, Michael and Filippone, Maurizio}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {4960--4968}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/heinonen25a/heinonen25a.pdf}, url = {https://proceedings.mlr.press/v258/heinonen25a.html}, abstract = {Introducing training-time augmentations is a key technique to enhance generalization and prepare deep neural networks against test-time corruptions. Inspired by the success of generative diffusion models, we propose a novel approach of coupling data mollification, in the form of image noising and blurring, with label smoothing to align predicted label confidences with image degradation. The method is simple to implement, introduces negligible overheads, and can be combined with existing augmentations. We demonstrate improved robustness and uncertainty quantification on the corrupted image benchmarks of CIFAR, TinyImageNet and ImageNet datasets.} }
Endnote
%0 Conference Paper %T Robust Classification by Coupling Data Mollification with Label Smoothing %A Markus Heinonen %A Ba-Hien Tran %A Michael Kampffmeyer %A Maurizio Filippone %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-heinonen25a %I PMLR %P 4960--4968 %U https://proceedings.mlr.press/v258/heinonen25a.html %V 258 %X Introducing training-time augmentations is a key technique to enhance generalization and prepare deep neural networks against test-time corruptions. Inspired by the success of generative diffusion models, we propose a novel approach of coupling data mollification, in the form of image noising and blurring, with label smoothing to align predicted label confidences with image degradation. The method is simple to implement, introduces negligible overheads, and can be combined with existing augmentations. We demonstrate improved robustness and uncertainty quantification on the corrupted image benchmarks of CIFAR, TinyImageNet and ImageNet datasets.
APA
Heinonen, M., Tran, B., Kampffmeyer, M. & Filippone, M.. (2025). Robust Classification by Coupling Data Mollification with Label Smoothing. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:4960-4968 Available from https://proceedings.mlr.press/v258/heinonen25a.html.

Related Material