Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 69 tok/s
Gemini 2.5 Pro 39 tok/s Pro
GPT-5 Medium 35 tok/s Pro
GPT-5 High 37 tok/s Pro
GPT-4o 103 tok/s Pro
Kimi K2 209 tok/s Pro
GPT OSS 120B 457 tok/s Pro
Claude Sonnet 4.5 34 tok/s Pro
2000 character limit reached

Fast Diffusion-Based Counterfactuals for Shortcut Removal and Generation (2312.14223v2)

Published 21 Dec 2023 in cs.CV

Abstract: Shortcut learning is when a model -- e.g. a cardiac disease classifier -- exploits correlations between the target label and a spurious shortcut feature, e.g. a pacemaker, to predict the target label based on the shortcut rather than real discriminative features. This is common in medical imaging, where treatment and clinical annotations correlate with disease labels, making them easy shortcuts to predict disease. We propose a novel detection and quantification of the impact of potential shortcut features via a fast diffusion-based counterfactual image generation that can synthetically remove or add shortcuts. Via a novel inpainting-based modification we spatially limit the changes made with no extra inference step, encouraging the removal of spatially constrained shortcut features while ensuring that the shortcut-free counterfactuals preserve their remaining image features to a high degree. Using these, we assess how shortcut features influence model predictions. This is enabled by our second contribution: An efficient diffusion-based counterfactual explanation method with significant inference speed-up at comparable image quality as state-of-the-art. We confirm this on two large chest X-ray datasets, a skin lesion dataset, and CelebA. Our code is publicly available at fastdime.compute.dtu.dk.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (73)
  1. Debugging tests for model explanations. In Advances in Neural Information Processing Systems, pages 700–712. Curran Associates, Inc., 2020.
  2. Post hoc explanations may be ineffective for detecting unknown spurious correlation. In International Conference on Learning Representations, 2022.
  3. Assessing the trustworthiness of saliency maps for localizing abnormalities in medical imaging. Radiology: Artificial Intelligence, 3(6), 2021.
  4. Diffusion visual counterfactual explanations. In Advances in Neural Information Processing Systems, pages 364–377. Curran Associates, Inc., 2022.
  5. Deep learning predicts hip fracture using confounding patient and healthcare variables. npj Digital Medicine, 2(1), 2019.
  6. Can we improve model robustness through secondary attribute counterfactuals? In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, 2021.
  7. Learning generative models using denoising density estimators. IEEE Transactions on Neural Networks and Learning Systems, 2023.
  8. Sparse Visual Counterfactual Explanations in Image Space, pages 133–148. Springer International Publishing, 2022.
  9. Detecting shortcut learning for fair medical AI using shortcut testing. Nature Communications, 14(1), 2023.
  10. Vggface2: A dataset for recognising faces across pose and age. In 2018 13th IEEE international conference on automatic face & gesture recognition (FG 2018), pages 67–74. IEEE, 2018.
  11. Towards robust classification model by counterfactual and invariant data generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15212–15221, 2021.
  12. StarGAN: Unified generative adversarial networks for multi-domain image-to-image translation. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, 2018.
  13. Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (ISIC). 2019a.
  14. Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (isic). arXiv preprint arXiv:1902.03368, 2019b.
  15. Gifsplanation via latent shift: A simple autoencoder approach to counterfactual generation for chest x-rays. In Medical Imaging with Deep Learning, 2021.
  16. Augmenting chest x-ray datasets with non-expert annotations. arXiv preprint arXiv:2309.02244, 2023.
  17. High fidelity image counterfactuals with probabilistic causal models. In Proceedings of the 40th International Conference on Machine Learning, pages 7390–7425. PMLR, 2023.
  18. AI for radiographic COVID-19 detection selects shortcuts over signal. Nature Machine Intelligence, 3(7):610–619, 2021.
  19. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009.
  20. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021.
  21. Diffeomorphic explanations with normalizing flows. In ICML Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models, 2021.
  22. medxgan: Visual explanations for medical classifiers through a generative latent space. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2936–2945, 2022.
  23. On the connection between adversarial robustness and saliency map interpretability. In Proceedings of the 36th International Conference on Machine Learning, pages 1823–1832. PMLR, 2019.
  24. Shortcut learning in deep neural networks. Nature Machine Intelligence, 2(11):665–673, 2020.
  25. AI recognition of patient race in medical imaging: a modelling study. The Lancet Digital Health, 4(6):e406–e414, 2022.
  26. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  27. GANs trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30, 2017.
  28. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020.
  29. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708, 2017.
  30. CheXpert: A large chest radiograph dataset with uncertainty labels and expert comparison. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01):590–597, 2019.
  31. Deep learning applied to chest x-rays: Exploiting and preventing shortcuts. In Proceedings of the 5th Machine Learning for Healthcare Conference, pages 750–782. PMLR, 2020.
  32. STEEX: steering counterfactual explanations with semantics. In European Conference on Computer Vision, pages 387–403. Springer, 2022.
  33. Diffusion models for counterfactual explanations. In Proceedings of the Asian Conference on Computer Vision, pages 858–876, 2022.
  34. Adversarial counterfactual visual explanations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16425–16435, 2023.
  35. Detecting shortcuts in medical images – a case study in chest x-rays. In International Symposium on Biomedical Imaging (ISBI), 2023.
  36. xGEMs: Generating examplars to explain black-box models. arXiv preprint arXiv:1806.08867, 2018.
  37. Cycle-consistent counterfactuals by latent transformations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10203–10212, 2022.
  38. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  39. Explaining in style: Training a gan to explain a classifier in stylespace. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 693–702, 2021.
  40. Achieving diversity in counterfactual explanations: a review and discussion. In 2023 ACM Conference on Fairness, Accountability, and Transparency. ACM, 2023.
  41. Deep learning face attributes in the wild. In Proceedings of International Conference on Computer Vision (ICCV), 2015.
  42. Stable bias: Analyzing societal representations in diffusion models. In NeurIPS, 2023.
  43. Repaint: Inpainting using denoising diffusion probabilistic models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11461–11471, 2022.
  44. GANterfactual—counterfactual explanations for medical non-experts using generative adversarial learning. Frontiers in Artificial Intelligence, 5, 2022.
  45. Shortcut detection with variational autoencoders. In ICML Workshop on Spurious Correlations,Invariance, and Stability, 2023.
  46. Uncovering and correcting shortcut learning in machine learning models for skin cancer diagnosis. Diagnostics, 12(1):40, 2021.
  47. Countergan: Generating counterfactuals for real-time recourse and interpretability using residual gans. In Uncertainty in Artificial Intelligence, pages 1488–1497. PMLR, 2022.
  48. Spurious features everywhere - large-scale detection of harmful spurious features in imagenet. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 20235–20246, 2023.
  49. Hidden stratification causes clinically meaningful failures in machine learning for medical imaging. In Proceedings of the ACM Conference on Health, Inference, and Learning. ACM, 2020a.
  50. Hidden stratification causes clinically meaningful failures in machine learning for medical imaging. In Proceedings of the ACM conference on health, inference, and learning, pages 151–159, 2020b.
  51. Reveal to revise: An explainable AI life cycle for iterative bias correction of deep models. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, pages 596–606. Springer Nature Switzerland, 2023.
  52. CIRCLe: Color invariant representation learning for unbiased classification of skin lesions. In Lecture Notes in Computer Science, pages 203–219. Springer Nature Switzerland, 2023.
  53. On counterfactual explanations under predictive multiplicity. In Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), pages 809–818. PMLR, 2020.
  54. Equitable modelling of brain imaging by counterfactual augmentation with morphologically constrained 3D deep generative models. Medical Image Analysis, 84:102723, 2023.
  55. Counterfactual interpolation augmentation (CIA): A unified approach to enhance fairness and explainability of DNN. In Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence Organization, 2022.
  56. Least squares estimation without priors or supervision. Neural computation, 23(2):374–420, 2011.
  57. Beyond trivial counterfactual explanations with diverse valuable explanations. In 2021 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE, 2021.
  58. Beyond trivial counterfactual explanations with diverse valuable explanations. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1056–1065, 2021.
  59. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, pages 234–241. Springer, 2015.
  60. Diffusion causal models for counterfactual estimation. In First Conference on Causal Learning and Reasoning, 2022.
  61. Image synthesis with a single (robust) classifier. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2019.
  62. Explanation by progressive exaggeration. In International Conference on Learning Representations, 2020.
  63. Explaining the black-box smoothly—a counterfactual approach. Medical Image Analysis, 84:102721, 2023.
  64. Counterfactual explanations can be manipulated. In Advances in Neural Information Processing Systems, pages 62–75. Curran Associates, Inc., 2021.
  65. Right for the wrong reason: Can interpretable ML techniques detect spurious correlations? In Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, pages 425–434. Springer Nature Switzerland, 2023.
  66. Training calibration-based counterfactual explainers for deep learning models in medical image analysis. Scientific Reports, 12(1), 2022.
  67. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data, 5(1), 2018.
  68. Diffusion-based visual counterfactual explanations–towards systematic quantitative evaluation. arXiv preprint arXiv:2308.06100, 2023.
  69. Kentaro Wada. Labelme: Image Polygonal Annotation with Python.
  70. ChestX-Ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2017.
  71. Association between surgical skin markings in dermoscopic images and diagnostic performance of a deep learning convolutional neural network for melanoma recognition. JAMA Dermatology, 155(10):1135, 2019.
  72. Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study. PLOS Medicine, 15(11):e1002683, 2018.
  73. Training certified detectives to track down the intrinsic shortcuts in COVID-19 chest x-ray data sets. Scientific Reports, 13(1), 2023.
Citations (7)
View on Semantic Scholar

Summary

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

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

Continue Learning

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up