# A simple way to make neural networks robust against diverse image corruptions

arXiv: https://arxiv.org/abs/2001.06057 · HTML: https://mlsanity.com/paper/2001.06057/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jan 2020.

## Built on by (most cited first, 30 of 30)

- [3D Common Corruptions and Data Augmentation](https://mlsanity.com/paper/2203.01441/) (2022)
- [Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformation](https://mlsanity.com/paper/2106.14999/) (2021)
- [RDumb: A simple approach that questions our progress in continual test-time adaptation](https://mlsanity.com/paper/2306.05401/) (2023)
- [How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review](https://mlsanity.com/paper/2107.12045/) (2021)
- [If your data distribution shifts, use self-learning](https://mlsanity.com/paper/2104.12928/) (2021)
- [Does enhanced shape bias improve neural network robustness to common corruptions?](https://mlsanity.com/paper/2104.09789/) (2021)
- [Large-scale Robustness Analysis of Video Action Recognition Models](https://mlsanity.com/paper/2207.01398/) (2022)
- [Robustness Analysis of Video-Language Models Against Visual and Language Perturbations](https://mlsanity.com/paper/2207.02159/) (2022)
- [Temporal Coherent Test-Time Optimization for Robust Video Classification](https://mlsanity.com/paper/2302.14309/) (2023)
- [IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifiers](https://mlsanity.com/paper/2310.04780/) (2023)
- [If a Human Can See It, So Should Your System: Reliability Requirements for Machine Vision Components](https://mlsanity.com/paper/2202.03930/) (2022)
- [Improving Model Robustness by Adaptively Correcting Perturbation Levels with Active Queries](https://mlsanity.com/paper/2103.14824/) (2021)
- [Foundation Model-oriented Robustness: Robust Image Model Evaluation with Pretrained Models](https://mlsanity.com/paper/2308.10632/) (2023)
- [Rethinking Streaming Machine Learning Evaluation](https://mlsanity.com/paper/2205.11473/) (2022)
- [Benchmarking Image Perturbations for Testing Automated Driving Assistance Systems](https://mlsanity.com/paper/2501.12269/) (2025)
- [A New Approach to Improve Learning-based Deepfake Detection in Realistic Conditions](https://mlsanity.com/paper/2203.11807/) (2022)
- [Natural Perturbed Training for General Robustness of Neural Network Classifiers](https://mlsanity.com/paper/2103.11372/) (2021)
- [LEAVES: Learning Views for Time-Series Biobehavioral Data in Contrastive Learning](https://mlsanity.com/paper/2210.07340/) (2022)
- [Robustness Enhancement in Neural Networks with Alpha-Stable Training Noise](https://mlsanity.com/paper/2311.10803/) (2023)
- [Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performance](https://mlsanity.com/paper/2402.19401/) (2024)
- [Nuisance-Label Supervision: Robustness Improvement by Free Labels](https://mlsanity.com/paper/2110.07118/) (2021)
- [Abutting Grating Illusion: Cognitive Challenge to Neural Network Models](https://mlsanity.com/paper/2208.03958/) (2022)
- [Robustness and invariance properties of image classifiers](https://mlsanity.com/paper/2209.02408/) (2022)
- [Explicitly Modeling Pre-Cortical Vision with a Neuro-Inspired Front-End Improves CNN Robustness](https://mlsanity.com/paper/2409.16838/) (2024)
- [Wiggling Weights to Improve the Robustness of Classifiers](https://mlsanity.com/paper/2111.09779/) (2021)
- [A Novel Framework for Assessment of Learning-based Detectors in Realistic Conditions with Application to Deepfake Detection](https://mlsanity.com/paper/2203.11797/) (2022)
- [Dynamic Batch Norm Statistics Update for Natural Robustness](https://mlsanity.com/paper/2310.20649/) (2023)
- [Improving Out-of-Distribution Data Handling and Corruption Resistance via Modern Hopfield Networks](https://mlsanity.com/paper/2408.11309/) (2024)
- [Growing Efficient Accurate and Robust Neural Networks on the Edge](https://mlsanity.com/paper/2410.07691/) (2024)
- [Deep Recurrence for Dynamical Segmentation Models](https://mlsanity.com/paper/2507.10143/) (2025)

## Abstract (arXiv)

The human visual system is remarkably robust against a wide range of naturally occurring variations and corruptions like rain or snow. In contrast, the performance of modern image recognition models strongly degrades when evaluated on previously unseen corruptions. Here, we demonstrate that a simple but properly tuned training with additive Gaussian and Speckle noise generalizes surprisingly well to unseen corruptions, easily reaching the previous state of the art on the corruption benchmark ImageNet-C (with ResNet50) and on MNIST-C. We build on top of these strong baseline results and show that an adversarial training of the recognition model against uncorrelated worst-case noise distributions leads to an additional increase in performance. This regularization can be combined with previously proposed defense methods for further improvement.