Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
113 tokens/sec
GPT-4o
12 tokens/sec
Gemini 2.5 Pro Pro
35 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
3 tokens/sec
DeepSeek R1 via Azure Pro
33 tokens/sec
2000 character limit reached

HarsanyiNet: Computing Accurate Shapley Values in a Single Forward Propagation (2304.01811v2)

Published 4 Apr 2023 in cs.LG, cs.AI, and cs.CV

Abstract: 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.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (57)
  1. Explaining deep neural networks with a polynomial time algorithm for shapley value approximation. In International Conference on Machine Learning, pp. 272–281. PMLR, 2019.
  2. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PloS one, 10(7):e0130140, 2015.
  3. Estimating or propagating gradients through stochastic neurons for conditional computation. arXiv preprint arXiv:1308.3432, 2013.
  4. Polynomial calculation of the shapley value based on sampling. Computers & Operations Research, 36(5):1726–1730, 2009.
  5. Algorithms to estimate shapley value feature attributions. arXiv preprint arXiv:2207.07605, 2022.
  6. L-shapley and c-shapley: Efficient model interpretation for structured data. International Conference on Learning Representation, 2019.
  7. Improving kernelshap: Practical shapley value estimation using linear regression. In International Conference on Artificial Intelligence and Statistics, pp.  3457–3465. PMLR, 2021.
  8. Explaining by removing: A unified framework for model explanation. Journal of Machine Learning Research, 2021.
  9. Understanding global feature contributions with additive importance measures. Advances in Neural Information Processing Systems, 33, 2020.
  10. Mutual information preserving back-propagation: Learn to invert for faithful attribution. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pp.  258–268, 2021.
  11. Discovering and explaining the representation bottleneck of dnns. International Conference on Learning Representation, 2022.
  12. UCI machine learning repository, 2017. URL http://archive.ics.uci.edu/ml.
  13. Interpretable explanations of black boxes by meaningful perturbation. In Proceedings of the IEEE International Conference on Computer Vision, pp.  3429–3437, 2017.
  14. Explaining and harnessing adversarial examples. In International Conference on Learning Representation, 2015.
  15. Grabisch, M. et al. Set functions, games and capacities in decision making, volume 46. Springer, 2016.
  16. Harsanyi, J. C. A simplified bargaining model for the n-person cooperative game. International Economic Review, 4(2):194–220, 1963.
  17. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  770–778, 2016.
  18. A benchmark for interpretability methods in deep neural networks. Advances in Neural Information Processing Systems, 32, 2019.
  19. Fastshap: Real-time shapley value estimation. In International Conference on Learning Representations, 2021.
  20. Efficient task-specific data valuation for nearest neighbor algorithms. In International Conference on Very Large Databases, 2019a.
  21. Towards efficient data valuation based on the shapley value. In International Conference on Artificial Intelligence and Statistics, 2019b.
  22. Scalability vs. utility: Do we have to sacrifice one for the other in data importance quantification? In IEEE Conference on Computer Vision and Pattern Recognition, 2021.
  23. Learning multiple layers of features from tiny images. 2009.
  24. Mnist handwritten digit database, 2010. URL http://yann.lecun.com/exdb/mnist/.
  25. Does a neural network really encode symbolic concepts? International Conference on Machine Learning, 2023.
  26. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 2017.
  27. Towards deep learning models resistant to adversarial attacks. In International Conference on Learning Representation, 2018.
  28. Sampling permutations for shapley value estimation. Journal of Machine Learning Research, 23:1–46, 2022.
  29. Explaining nonlinear classification decisions with deep taylor decomposition. Pattern recognition, 65:211–222, 2017.
  30. Automated flower classification over a large number of classes. In Indian Conference on Computer Vision, Graphics & Image Processing, 2008.
  31. A multilinear sampling algorithm to estimate shapley values. International Conference on Pattern Recognition, 2021.
  32. Covid-net cxr-2: An enhanced deep convolutional neural network design for detection of covid-19 cases from chest x-ray images. In Front Med (Lausanne), 2022.
  33. Model agnostic supervised local explanations. Advances in Neural Information Processing Systems, 31, 2018.
  34. Defining and quantifying the emergence of sparse concepts in dnns. IEEE Conference on Computer Vision and Pattern Recognition, 2023a.
  35. Where we have arrived in proving the emergence of sparse symbolic concepts in ai models. arXiv preprint arXiv:2305.01939, 2023b.
  36. ”why should i trust you?” explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.  1135–1144, 2016.
  37. Restricting the flow: Information bottlenecks for attribution. International Conference on Learning Representation, 2020.
  38. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision, pp.  618–626, 2017.
  39. Shapley, L. S. A value for n-person games. In Contributions to the Theory of Games, 2(28):307–317, 1953.
  40. Not just a black box: Learning important features through propagating activation differences. arXiv preprint arXiv:1605.01713, 2016.
  41. Learning important features through propagating activation differences. In International Conference on Machine Learning, pp. 3145–3153. PMLR, 2017.
  42. A projected stochastic gradient algorithm for estimating shapley value applied in attribute importance. In International Cross-Domain Conference on Machine Learning and Knowledge Extraction, 2020.
  43. Very deep convolutional networks for large-scale image recognition. In International Conference on Learning Representation, 2015.
  44. Deep inside convolutional networks: Visualising image classification models and saliency maps. International Conference on Learning Representations, 2014.
  45. Striving for simplicity: The all convolutional net. International Conference on Learning Representations, 2015.
  46. An efficient explanation of individual classifications using game theory. The Journal of Machine Learning Research, 11:1–18, 2010.
  47. The many shapley values for model explanation. In International Conference on Machine Learning, pp. 9269–9278. PMLR, 2020.
  48. Axiomatic attribution for deep networks. In International Conference on Machine Learning, pp. 3319–3328. PMLR, 2017.
  49. Resmlp: Feedforward networks for image classification with data-efficient training. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022.
  50. Accelerating shapley explanation via contributive cooperator selection. In International Conference on Machine Learning, pp. 22576–22590. PMLR, 2022.
  51. Covid-net: a tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray image. In Scientific Reports, 2020.
  52. Shapley explanation networks. In International Conference on Learning Representations, 2021.
  53. Weber, R. J. Probabilistic values for games. The Shapley Value. Essays in Honor of Lloyd S. Shapley, 101–119, 1988.
  54. Resnet strikes back: An improved training procedure in timm. In Neural Information Processing Systems, 2021.
  55. Young, H. P. Monotonic solutions of cooperative games. International Journal of Game Theory, 14:65–72, 1985.
  56. Learning deep features for discriminative localization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  2921–2929, 2016.
  57. Visualizing deep neural network decisions: Prediction difference analysis. International Conference on Learning Representations, 2017.
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now