mlsanityData to 7 Oct 2026

A New Framework to Analyse the Distributional Robustness of Deep Neural Networks

arXiv 2605.21313 ↗First posted 20 May 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep neural networks have achieved impressive performance on a variety of tasks, but their brittleness to distributional shifts remains a significant barrier to real-world deployment. In this paper, we propose a framework to analyse and quantify the distributional robustness of neural networks by studying the interactions between layer weights and activations. We model these interactions using Bernoulli distributions, using the separation between classes as a diagnostic proxy for robustness. We demonstrate the usefulness of this framework through models trained on CIFAR-10 and ImageNet. We show that our proposed metrics can distinguish between networks that have memorised their training data and those that have not. We also perform analogous experiments in the activation space and find that the same properties do not hold up. Additionally, we investigate the behaviour of our metrics under various distribution shifts and show that these shifts reduce separation under our path-based diagnostics. Our results suggest that this framework provides useful model-level diagnostics of representation structure and robustness.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 28 of its 32 references that are in the map. First posted on arXiv 20 May 2026.

Builds on

28 of 32 references.

References · 28

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  6. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  7. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  8. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  9. 2013Intriguing properties of neural networkscited by 5,114
  10. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  11. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  12. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
16 more
  1. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  2. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  3. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  4. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  5. 2018Spectral Signatures in Backdoor Attackscited by 457
  6. 2018Certified Defenses against Adversarial Examplescited by 305
  7. 2018Do CIFAR-10 Classifiers Generalize to CIFAR-10?cited by 197
  8. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 123
  9. 2018Are adversarial examples inevitable?cited by 113
  10. 2019Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizationscited by 72
  11. 2021This looks more like that: Enhancing Self-Explaining Models by Prototypical Relevance Propagationcited by 43
  12. 2022Detection of out-of-distribution samples using binary neuron activation patternscited by 15
  13. 2024Gradient-Regularized Out-of-Distribution Detectioncited by 13
  14. 2023Prototypical Self-Explainable Models Without Re-trainingcited by 6
  15. 2019Detection and Mitigation of Rare Subclasses in Deep Neural Network Classifierscited by 4
  16. 2020Explaining Predictions of Deep Neural Classifier via Activation Analysiscited by 3

Built on by

Not cited yet.