mlsanityData to 14 Oct 2026

Attribution for Enhanced Explanation with Transferable Adversarial eXploration

arXiv 2412.19523 ↗First posted 27 Dec 2024cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

The interpretability of deep neural networks is crucial for understanding model decisions in various applications, including computer vision. AttEXplore++, an advanced framework built upon AttEXplore, enhances attribution by incorporating transferable adversarial attack methods such as MIG and GRA, significantly improving the accuracy and robustness of model explanations. We conduct extensive experiments on five models, including CNNs (Inception-v3, ResNet-50, VGG16) and vision transformers (MaxViT-T, ViT-B/16), using the ImageNet dataset. Our method achieves an average performance improvement of 7.57% over AttEXplore and 32.62% compared to other state-of-the-art interpretability algorithms. Using insertion and deletion scores as evaluation metrics, we show that adversarial transferability plays a vital role in enhancing attribution results. Furthermore, we explore the impact of randomness, perturbation rate, noise amplitude, and diversity probability on attribution performance, demonstrating that AttEXplore++ provides more stable and reliable explanations across various models. We release our code at: https://anonymous.4open.science/r/ATTEXPLOREP-8435/

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 36 of its 54 references that are in the map. First posted on arXiv 27 Dec 2024.

Builds on

36 of 54 references.

References · 36

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  5. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  8. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  9. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  10. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  11. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  12. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
24 more
  1. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  2. 2016Adversarial examples in the physical worldcited by 1,769
  3. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  4. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  5. 2017SmoothGrad: removing noise by adding noisecited by 1,109
  6. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 838
  7. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 599
  8. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 476
  9. 2019Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attackscited by 442
  10. 2019A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAIcited by 404
  11. 2019Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attackscited by 355
  12. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 345
  13. 2017Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradientscited by 236
  14. 2022Frequency Domain Model Augmentation for Adversarial Attackcited by 130
  15. 2021Guided Integrated Gradients: An Adaptive Path Method for Removing Noisecited by 102
  16. 2022Improving Adversarial Transferability via Neuron Attribution-Based Attackscited by 93
  17. 2023Structure Invariant Transformation for better Adversarial Transferabilitycited by 62
  18. 2021Fast Axiomatic Attribution for Neural Networkscited by 35
  19. 2021Robust Models Are More Interpretable Because Attributions Look Normalcited by 21
  20. 2021DI-AA: An Interpretable White-box Attack for Fooling Deep Neural Networkscited by 10
  21. 2023MFABA: A More Faithful and Accelerated Boundary-based Attribution Method for Deep Neural Networkscited by 7
  22. 2024Benchmarking Transferable Adversarial Attackscited by 5
  23. 2022Interpretability of Machine Learning Methods Applied to Neuroimagingcited by 4
  24. 2024Enhancing Model Interpretability with Local Attribution over Global Explorationcited by 4

Built on by

Not cited yet.