mlsanityData to 7 Oct 2026

Neuron Abandoning Attention Flow: Visual Explanation of Dynamics inside CNN Models

arXiv 2412.01202 ↗First posted 2 Dec 2024cs.CV cs.AI cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

In this paper, we present a Neuron Abandoning Attention Flow (NAFlow) method to address the open problem of visually explaining the attention evolution dynamics inside CNNs when making their classification decisions. A novel cascading neuron abandoning back-propagation algorithm is designed to trace neurons in all layers of a CNN that involve in making its prediction to address the problem of significant interference from abandoned neurons. Firstly, a Neuron Abandoning Back-Propagation (NA-BP) module is proposed to generate Back-Propagated Feature Maps (BPFM) by using the inverse function of the intermediate layers of CNN models, on which the neurons not used for decision-making are abandoned. Meanwhile, the cascading NA-BP modules calculate the tensors of importance coefficients which are linearly combined with the tensors of BPFMs to form the NAFlow. Secondly, to be able to visualize attention flow for similarity metric-based CNN models, a new channel contribution weights module is proposed to calculate the importance coefficients via Jacobian Matrix. The effectiveness of the proposed NAFlow is validated on nine widely-used CNN models for various tasks of general image classification, contrastive learning classification, few-shot image classification, and image retrieval.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 37 of its 65 references that are in the map. First posted on arXiv 2 Dec 2024.

Builds on

37 of 65 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  5. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  7. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  8. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  9. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  10. 2016Matching Networks for One Shot Learningcited by 2,246
  11. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  12. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
25 more
  1. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
  2. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  3. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  4. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  5. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
  6. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 825
  7. 2020Transformer Interpretability Beyond Attention Visualizationcited by 620
  8. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 611
  9. 2019Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networkscited by 575
  10. 2019Multi-Similarity Loss with General Pair Weighting for Deep Metric Learningcited by 368
  11. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  12. 2020An Investigation of Why Overparameterization Exacerbates Spurious Correlationscited by 310
  13. 2019Full-Gradient Representation for Neural Network Visualizationcited by 189
  14. 2020Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learningcited by 179
  15. 2018BAM: Bottleneck Attention Modulecited by 171
  16. 2018Adversarial Complementary Learning for Weakly Supervised Object Localizationcited by 166
  17. 2019FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inferencecited by 162
  18. 2020Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNscited by 151
  19. 2022On Feature Learning in the Presence of Spurious Correlationscited by 145
  20. 2018Attention Branch Network: Learning of Attention Mechanism for Visual Explanationcited by 108
  21. 2021A Comprehensive Taxonomy for Explainable Artificial Intelligence: A Systematic Survey of Surveys on Methods and Conceptscited by 90
  22. 2020Discrimination-aware Network Pruning for Deep Model Compressioncited by 50
  23. 2021Towards Better Explanations of Class Activation Mappingcited by 47
  24. 2022A Theoretical Analysis of the Learning Dynamics under Class Imbalancecited by 20
  25. 2021Content-Aware Convolutional Neural Networkscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Adaptive receptive field-based spatial-frequency feature reconstruction network for fine-grained few-shot image classificationcited by 2
  2. 2026GloResNet: A lightweight 3D CNN with global topological features for preterm brain injury predictioncited by 1
  3. 2026Phase-Aware Spatial-Frequency Fusion for Few-Shot Fine-Grained Image Classificationcited by 0