Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
194 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Deep Backtracking Counterfactuals for Causally Compliant Explanations (2310.07665v4)

Published 11 Oct 2023 in cs.AI, cs.LG, and stat.ML

Abstract: Counterfactuals answer questions of what would have been observed under altered circumstances and can therefore offer valuable insights. Whereas the classical interventional interpretation of counterfactuals has been studied extensively, backtracking constitutes a less studied alternative where all causal laws are kept intact. In the present work, we introduce a practical method called deep backtracking counterfactuals (DeepBC) for computing backtracking counterfactuals in structural causal models that consist of deep generative components. We propose two distinct versions of our method--one utilizing Langevin Monte Carlo sampling and the other employing constrained optimization--to generate counterfactuals for high-dimensional data. As a special case, our formulation reduces to methods in the field of counterfactual explanations. Compared to these, our approach represents a causally compliant, versatile and modular alternative. We demonstrate these properties experimentally on a modified version of MNIST and CelebA.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (67)
  1. Causalworld: A robotic manipulation benchmark for causal structure and transfer learning. International Conference on Learning Representations, 2021.
  2. On Pearl’s Hierarchy and the Foundations of Causal Inference. Probabilistic and Causal Inference: The Works of Judea Pearl, pp.  507–556, 2022.
  3. Sander Beckers. Causal Explanations and XAI. In Conference on Causal Learning and Reasoning, pp.  90–109, 2022.
  4. Sparse Visual Counterfactual Explanations in Image Space. In DAGM German Conference on Pattern Recognition, pp.  133–148. Springer, 2022.
  5. Learning Linear Causal Representations from Interventions under General Nonlinear Mixing. arXiv preprint arXiv:2306.02235, 2023.
  6. A New Paradigm for Counterfactual Reasoning in Fairness and Recourse. arXiv preprint arXiv:2401.13935, 2024.
  7. Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learning. Journal of Machine Learning Research, 20(178):1–29, 2019.
  8. Evaluating and Mitigating Bias in Image Classifiers: A Causal Perspective Using Counterfactuals. In Winter Conference on Applications of Computer Vision, pp.  915–924, 2022.
  9. Neural Spline Flows. Advances in Neural Information Processing Systems, 32:7509–7520, 2019.
  10. Falcon, William and The PyTorch Lightning team. PyTorch Lightning, 2019.
  11. Back on track: Backtracking in counterfactual reasoning. In Annual Meeting of the Cognitive Science Society, volume 35, pp.  2386–2391, 2013.
  12. From Variational to Deterministic Autoencoders. In International Conference on Learning Representations, 2020.
  13. Generative Adversarial Nets. Advances in Neural Information Processing Systems, 27:53–65, 2014.
  14. Learning Functional Causal Models with Generative Neural Networks. Explainable and Interpretable Models in Computer Vision and Machine Learning, pp.  39–80, 2018.
  15. Counterfactual Visual Explanations. In International Conference on Machine Learning, pp.  2376–2384, 2019.
  16. Riccardo Guidotti. Counterfactual explanations and how to find them: literature review and benchmarking. Data Mining and Knowledge Discovery, pp.  1–55, 2022.
  17. β𝛽\betaitalic_β-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework. In International Conference on Learning Representations, 2017.
  18. Denoising Diffusion Probabilistic Models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020.
  19. A Causal Lens for Controllable Text Generation. Advances in Neural Information Processing Systems, 34:24941–24955, 2021.
  20. Identifiability of latent-variable and structural-equation models: from linear to nonlinear. arXiv preprint arXiv:2302.02672, 2023.
  21. STEEX: Steering Counterfactual Explanations with Semantics. In European Conference on Computer Vision, pp.  387–403, 2022.
  22. Categorical Reparameterization with Gumbel-Softmax. International Conference on Learning Representations, 2017.
  23. Causal normalizing flows: from theory to practice. Advances in Neural Information Processing Systems, 2023.
  24. Algorithmic recourse under imperfect causal knowledge: a probabilistic approach. Advances in Neural Information Processing Systems, 33:265–277, 2020.
  25. Algorithmic Recourse: from Counterfactual Explanations to Interventions. In ACM Conference on Fairness, Accountability, and Transparency, pp.  353–362, 2021.
  26. A Survey of Algorithmic Recourse: Contrastive Explanations and Consequential Recommendations. ACM Computing Surveys, 55(5):1–29, 2022.
  27. Causal Autoregressive Flows. In International Conference on Artificial Intelligence and Statistics, pp.  3520–3528, 2021.
  28. Justin Khoo. Backtracking Counterfactuals Revisited. Mind, 126(503):841–910, 2017.
  29. Auto-Encoding Variational Bayes. International Conference on Learning Representations, 2014.
  30. CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training. International Conference on Learning Representations, 2018.
  31. Generating Sparse Counterfactual Explanations for Multivariate Time Series. arXiv preprint arXiv:2206.00931, 2022.
  32. David Lewis. Counterfactual Dependence and Time’s Arrow. Noûs, pp.  455–476, 1979.
  33. Causal Component Analysis. Advances in Neural Information Processing Systems, 2023.
  34. BISCUIT: Causal Representation Learning from Binary Interactions. In Conference on Uncertainty in Artificial Intelligence, 2023.
  35. Deep Learning Face Attributes in the Wild. In International Conference on Computer Vision, pp.  3730–3738, 2015.
  36. Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations. In International Conference on Machine Learning, pp.  4114–4124, 2019.
  37. An Improved Probabilistic Account of Counterfactual Reasoning. Psychological Review, 122(4):700, 2015.
  38. Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers. arXiv preprint arXiv:1912.03277, 2019.
  39. Jorge J Moré. The Levenberg-Marquardt algorithm: implementation and theory. In Numerical Analysis: Proceedings of the Biennial Conference Held at Dundee, June 28–July 1, 1977, pp.  105–116. Springer, 2006.
  40. Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations. In ACM Conference on Fairness, Accountability, and Transparency, pp.  607–617, 2020.
  41. Counterfactual Identifiability of Bijective Causal Models. arXiv preprint arXiv:2302.02228, 2023.
  42. Giorgio Parisi. Correlation Functions and Computer Simulations. Nuclear Physics B, 180(3):378–384, 1981.
  43. PyTorch: An Imperative Style, High-Performance Deep Learning Library. Advances in Neural Information Processing Systems, 32:8024–8035, 2019.
  44. Deep Structural Causal Models for Tractable Counterfactual Inference. Advances in Neural Information Processing Systems, 33:857–869, 2020.
  45. Judea Pearl. Causality. Cambridge University Press, 2009.
  46. Causal Discovery in Heterogeneous Environments Under the Sparse Mechanism Shift Hypothesis. Advances in Neural Information Processing Systems, 35:10904–10917, 2022.
  47. Embrace the Gap: VAEs Perform Independent Mechanism Analysis. Advances in Neural Information Processing Systems, 35:12040–12057, 2022.
  48. Variational Inference with Normalizing Flows. In International Conference on Machine Learning, pp.  1530–1538, 2015.
  49. Lance J. Rips. Two Causal Theories of Counterfactual Conditionals. Cognitive Science, 34(2):175–221, 2010.
  50. Beyond Trivial Counterfactual Explanations with Diverse Valuable Explanations. In International Conference on Computer Vision, pp.  1056–1065, 2021.
  51. Neal J. Roese. Counterfactual Thinking. Psychological Bulletin, 121(1):133–148, 1997.
  52. Diffusion Causal Models for Counterfactual Estimation. Conference on Causal Learning and Reasoning, pp.  1–21, 2022.
  53. VACA: Design of Variational Graph Autoencoders for Interventional and Counterfactual Queries. AAAI Conference on Artificial Intelligence, pp.  8159–8168, 2022.
  54. Timothy Sauer. Computational Solution of Stochastic Differential Equations. Wiley Interdisciplinary Reviews: Computational Statistics, 5(5):362–371, 2013.
  55. Toward Causal Representation Learning. IEEE, 109(5):612–634, 2021.
  56. Learning Structured Output Representation using Deep Conditional Generative Models. Advances in Neural Information Processing Systems, 28:3483–3491, 2015.
  57. Peter Spirtes. Introduction to Causal Inference. Journal of Machine Learning Research, 11(5):1643–1662, 2010.
  58. normflows: A PyTorch Package for Normalizing Flows. Journal of Open Source Software, 8(86):5361, 2023.
  59. Uk biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS medicine, 12(3):e1001779, 2015.
  60. Causal Generative Explainers using Counterfactual Inference: A Case Study on the Morpho-MNIST Dataset. arXiv preprint arXiv:2401.11394, 2024.
  61. Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review. arXiv preprint arXiv:2010.10596, 2020.
  62. Nonparametric Identifiability of Causal Representations from Unknown Interventions. Advances in Neural Information Processing Systems, 2023a.
  63. Backtracking Counterfactuals. In Conference on Causal Learning and Reasoning, pp.  177–196, 2023b.
  64. Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR. Harvard Journal of Law & Technology, 31:841–887, 2017.
  65. Learning Likelihoods with Conditional Normalizing Flows. arXiv preprint arXiv:1912.00042, 2019.
  66. Multi-domain translation between single-cell imaging and sequencing data using autoencoders. Nature Communications, 12(1):31, 2021a.
  67. CausalVAE: Structured Causal Disentanglement in Variational Autoencoder. In Conference on Computer Vision and Pattern Recognition, pp.  9593–9602, 2021b.
Citations (3)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets