mlsanityData to 14 Oct 2026

Path Choice Matters for Clear Attribution in Path Methods

arXiv 2401.10442 ↗First posted 19 Jan 2024cs.CV cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Rigorousness and clarity are both essential for interpretations of DNNs to engender human trust. Path methods are commonly employed to generate rigorous attributions that satisfy three axioms. However, the meaning of attributions remains ambiguous due to distinct path choices. To address the ambiguity, we introduce Concentration Principle, which centrally allocates high attributions to indispensable features, thereby endowing aesthetic and sparsity. We then present SAMP, a model-agnostic interpreter, which efficiently searches the near-optimal path from a pre-defined set of manipulation paths. Moreover, we propose the infinitesimal constraint (IC) and momentum strategy (MS) to improve the rigorousness and optimality. Visualizations show that SAMP can precisely reveal DNNs by pinpointing salient image pixels. We also perform quantitative experiments and observe that our method significantly outperforms the counterparts. Code: https://github.com/zbr17/SAMP.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 35 of its 44 references that are in the map. First posted on arXiv 19 Jan 2024.

Builds on

35 of 44 references.

References · 35

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  4. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  5. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  7. 2015Learning Deep Features for Discriminative Localizationcited by 2,436
  8. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  10. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  11. 2017SmoothGrad: removing noise by adding noisecited by 1,109
  12. 2014Striving for Simplicity: The All Convolutional Netcited by 990
23 more
  1. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 939
  2. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 838
  3. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 612
  4. 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
  5. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 345
  6. 2016Top-down Neural Attention by Excitation Backpropcited by 321
  7. 2020Neural Additive Models: Interpretable Machine Learning with Neural Netscited by 319
  8. 2022From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AIcited by 302
  9. 2016Not Just a Black Box: Learning Important Features Through Propagating Activation Differencescited by 297
  10. 2017Visualizing Deep Neural Network Decisions: Prediction Difference Analysiscited by 166
  11. 2017Interpretable Convolutional Neural Networkscited by 162
  12. 2018Visual Interpretability for Deep Learning: a Surveycited by 155
  13. 2021Benchmarking and Survey of Explanation Methods for Black Box Modelscited by 136
  14. 2019XRAI: Better Attributions Through Regionscited by 116
  15. 2021Guided Integrated Gradients: An Adaptive Path Method for Removing Noisecited by 102
  16. 2020Attribution in Scale and Spacecited by 61
  17. 2021Entropy-based Logic Explanations of Neural Networkscited by 51
  18. 2020NBDT: Neural-Backed Decision Treescited by 46
  19. 2022Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measurecited by 33
  20. 2020Training Interpretable Convolutional Neural Networks by Differentiating Class-specific Filterscited by 26
  21. 2021Self-Interpretable Model with TransformationEquivariant Interpretationcited by 21
  22. 2021Interpretable Compositional Convolutional Neural Networkscited by 16
  23. 2020DANCE: Enhancing saliency maps using decoyscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 3.

4 papers.

  1. 2024Efficient and Accurate Explanation Estimation with Distribution Compressioncited by 4
  2. 2026Manifold-Aligned Guided Integrated Gradients for Reliable Feature Attributioncited by 2
  3. 2026Diffusion Integrated Gradients: Controllable Path Generation for Flexible Feature Attributioncited by 1
  4. 2026Consistent Evidence, Robust Recognition: Faithful Attribution Regularization under Geometric Transformationscited by 0