mlsanityData to 7 Oct 2026

Technical Note: Defining and Quantifying AND-OR Interactions for Faithful and Concise Explanation of DNNs

arXiv 2304.13312 ↗First posted 26 Apr 2023cs.LG cs.AI cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

In this technical note, we aim to explain a deep neural network (DNN) by quantifying the encoded interactions between input variables, which reflects the DNN's inference logic. Specifically, we first rethink the definition of interactions, and then formally define faithfulness and conciseness for interaction-based explanation. To this end, we propose two kinds of interactions, i.e., the AND interaction and the OR interaction. For faithfulness, we prove the uniqueness of the AND (OR) interaction in quantifying the effect of the AND (OR) relationship between input variables. Besides, based on AND-OR interactions, we design techniques to boost the conciseness of the explanation, while not hurting the faithfulness. In this way, the inference logic of a DNN can be faithfully and concisely explained by a set of symbolic concepts.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 24 of its 31 references that are in the map. First posted on arXiv 26 Apr 2023.

Builds on

24 of 31 references.

References · 24

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  5. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  6. 2015Understanding Neural Networks Through Deep Visualizationcited by 332
  7. 2017Real Time Image Saliency for Black Box Classifierscited by 246
  8. 2019The Shapley Taylor Interaction Indexcited by 143
  9. 2019Explaining Deep Neural Networks with a Polynomial Time Algorithm for Shapley Values Approximationcited by 115
  10. 2020A Unified Approach to Interpreting and Boosting Adversarial Transferabilitycited by 71
  11. 2022Faith-Shap: The Faithful Shapley Interaction Indexcited by 68
  12. 2021Discovering and Explaining the Representation Bottleneck of DNNscited by 42
12 more
  1. 2021Defining and Quantifying the Emergence of Sparse Concepts in DNNscited by 36
  2. 2020Interpreting and Boosting Dropout from a Game-Theoretic Viewcited by 34
  3. 2023Does a Neural Network Really Encode Symbolic Concepts?cited by 27
  4. 2023Explaining Generalization Power of a DNN Using Interactive Conceptscited by 25
  5. 2021A Game-Theoretic Taxonomy of Visual Concepts in DNNscited by 16
  6. 2023HarsanyiNet: Computing Accurate Shapley Values in a Single Forward Propagationcited by 16
  7. 2020A Unified Taylor Framework for Revisiting Attribution Methodscited by 13
  8. 2021Interpreting Attributions and Interactions of Adversarial Attackscited by 13
  9. 2022Proving Common Mechanisms Shared by Twelve Methods of Boosting Adversarial Transferabilitycited by 12
  10. 2023Understanding and Unifying Fourteen Attribution Methods with Taylor Interactionscited by 11
  11. 2023Can the Inference Logic of Large Language Models be Disentangled into Symbolic Concepts?cited by 11
  12. 2021A Hypothesis for the Aesthetic Appreciation in Neural Networkscited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 5; 2025: 2; 2026: 1.

10 papers.

  1. 2025ProxySPEX: Inference-Efficient Interpretability via Sparse Feature Interactions in LLMscited by 16
  2. 2024Defining and Extracting generalizable interaction primitives from DNNscited by 14
  3. 2024Towards the Dynamics of a DNN Learning Symbolic Interactionscited by 13
  4. 2023Where We Have Arrived in Proving the Emergence of Sparse Symbolic Concepts in AI Modelscited by 8
  5. 2024Two-Phase Dynamics of Interactions Explains the Starting Point of a DNN Learning Over-Fitted Featurescited by 5
  6. 2024Layerwise Change of Knowledge in Neural Networkscited by 4
  7. 2024Quantifying In-Context Reasoning Effects and Memorization Effects in LLMscited by 3
  8. 2023Towards Attributions of Input Variables in a Coalitioncited by 1
  9. 2025Revisiting Generalization Power of a DNN in Terms of Symbolic Interactionscited by 1
  10. 2026Reconciling Contradictory Views on the Effectiveness of SFT in LLMs: An Interaction Perspectivecited by 0