mlsanityData to 7 Oct 2026

HarsanyiNet: Computing Accurate Shapley Values in a Single Forward Propagation

arXiv 2304.01811 ↗First posted 4 Apr 2023cs.LG cs.AI cs.CVcited by 16

Not on the map. Why?

Abstract (arXiv)

The Shapley value is widely regarded as a trustworthy attribution metric. However, when people use Shapley values to explain the attribution of input variables of a deep neural network (DNN), it usually requires a very high computational cost to approximate relatively accurate Shapley values in real-world applications. Therefore, we propose a novel network architecture, the HarsanyiNet, which makes inferences on the input sample and simultaneously computes the exact Shapley values of the input variables in a single forward propagation. The HarsanyiNet is designed on the theoretical foundation that the Shapley value can be reformulated as the redistribution of Harsanyi interactions encoded by the network.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 40 of its 60 references that are in the map. First posted on arXiv 4 Apr 2023.

Builds on

40 of 60 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  5. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  7. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  8. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  9. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  10. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  11. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  12. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
28 more
  1. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  2. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 611
  3. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  4. 2021ResNet strikes back: An improved training procedure in timmcited by 365
  5. 2019The many Shapley values for model explanationcited by 341
  6. 2021ResMLP: Feedforward networks for image classification with data-efficient trainingcited by 338
  7. 2016Not Just a Black Box: Learning Important Features Through Propagating Activation Differencescited by 295
  8. 2019Towards Efficient Data Valuation Based on the Shapley Valuecited by 279
  9. 2020COVID-Net: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest X-Ray Imagescited by 264
  10. 2020Explaining by Removing: A Unified Framework for Model Explanationcited by 218
  11. 2017Visualizing Deep Neural Network Decisions: Prediction Difference Analysiscited by 166
  12. 2020Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regressioncited by 153
  13. 2019Efficient Task-Specific Data Valuation for Nearest Neighbor Algorithmscited by 147
  14. 2021FastSHAP: Real-Time Shapley Value Estimationcited by 123
  15. 2019Explaining Deep Neural Networks with a Polynomial Time Algorithm for Shapley Values Approximationcited by 115
  16. 2022Algorithms to estimate Shapley value feature attributionscited by 114
  17. 2020Restricting the Flow: Information Bottlenecks for Attributioncited by 110
  18. 2021Sampling Permutations for Shapley Value Estimationcited by 101
  19. 2018L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Datacited by 100
  20. 2018Model Agnostic Supervised Local Explanationscited by 83
  21. 2021Discovering and Explaining the Representation Bottleneck of DNNscited by 42
  22. 2021Shapley Explanation Networkscited by 41
  23. 2021Defining and Quantifying the Emergence of Sparse Concepts in DNNscited by 36
  24. 2020A Multilinear Sampling Algorithm to Estimate Shapley Valuescited by 33
  25. 2023Does a Neural Network Really Encode Symbolic Concepts?cited by 27
  26. 2021COVID-Net CXR-2: An Enhanced Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest X-ray Imagescited by 22
  27. 2022Accelerating Shapley Explanation via Contributive Cooperator Selectioncited by 13
  28. 2021Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attributioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 2; 2025: 8; 2026: 3.

16 papers.

Sort:

By area:

  1. 2023Technical Note: Defining and Quantifying AND-OR Interactions for Faithful and Concise Explanation of DNNscited by 10
  2. 2025INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Samplingcited by 10
  3. 2025SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelismcited by 8
  4. 2024Diagnosing the Compositional Knowledge of Vision Language Models from a Game-Theoretic Viewcited by 7
  5. 2025Generalized Attention Flow: Feature Attribution for Transformer Models via Maximum Flowcited by 3
  6. 2025Prediction via Shapley Value Regressioncited by 3
  7. 2023Fast Shapley Value Estimation: A Unified Approachcited by 2
  8. 2025A Survey on Interpretability in Visual Recognitioncited by 2
  9. 2025Enhancing Interpretability for Vision Models via Shapley Value Optimizationcited by 2
  10. 2023TVE: Learning Meta-attribution for Transferable Vision Explainercited by 1
  11. 2024Is Data Valuation Learnable and Interpretable?cited by 1
  12. 2025Composing Linear Layers from Irreduciblescited by 1
4 more
  1. 2026PhaseWin: An Efficient Search Algorithm for Faithful Visual Attributioncited by 1
  2. 2025Beyond Models! Explainable Data Valuation and Metric Adaption for Recommendationcited by 0
  3. 2026More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffoldingcited by 0
  4. 2026Verified SHAP: Provable Bounds for Exact Shapley Values of Neural Networkscited by 0