Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 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

Counterfactual Fairness by Combining Factual and Counterfactual Predictions (2409.01977v3)

Published 3 Sep 2024 in cs.LG

Abstract: In high-stake domains such as healthcare and hiring, the role of ML in decision-making raises significant fairness concerns. This work focuses on Counterfactual Fairness (CF), which posits that an ML model's outcome on any individual should remain unchanged if they had belonged to a different demographic group. Previous works have proposed methods that guarantee CF. Notwithstanding, their effects on the model's predictive performance remains largely unclear. To fill in this gap, we provide a theoretical study on the inherent trade-off between CF and predictive performance in a model-agnostic manner. We first propose a simple but effective method to cast an optimal but potentially unfair predictor into a fair one without losing the optimality. By analyzing its excess risk in order to achieve CF, we quantify this inherent trade-off. Further analysis on our method's performance with access to only incomplete causal knowledge is also conducted. Built upon it, we propose a performant algorithm that can be applied in such scenarios. Experiments on both synthetic and semi-synthetic datasets demonstrate the validity of our analysis and methods.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. Causal context connects counterfactual fairness to robust prediction and group fairness. Advances in Neural Information Processing Systems, 36, 2024.
  2. Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in neural information processing systems, 29, 2016.
  3. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258, 2021.
  4. Adrienne Brackey. Analysis of Racial Bias in Northpointe’s COMPAS Algorithm. PhD thesis, Tulane University School of Science and Engineering, 2019.
  5. Evaluating the predictive validity of the compas risk and needs assessment system. Criminal Justice and behavior, 36(1):21–40, 2009.
  6. Three naive bayes approaches for discrimination-free classification. Data mining and knowledge discovery, 21:277–292, 2010.
  7. Why is my classifier discriminatory? Advances in neural information processing systems, 31, 2018.
  8. Silvia Chiappa. Path-specific counterfactual fairness. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 7801–7808, 2019.
  9. David Maxwell Chickering. Optimal structure identification with greedy search. Journal of machine learning research, 3(Nov):507–554, 2002.
  10. Fair regression with wasserstein barycenters. Advances in Neural Information Processing Systems, 33:7321–7331, 2020.
  11. Order-independent constraint-based causal structure learning. J. Mach. Learn. Res., 15(1):3741–3782, 2014.
  12. The measure and mismeasure of fairness. The Journal of Machine Learning Research, 24(1):14730–14846, 2023.
  13. Disparities in dermatology ai: assessments using diverse clinical images. arXiv preprint arXiv:2111.08006, 2021.
  14. Empirical risk minimization under fairness constraints. Advances in neural information processing systems, 31, 2018.
  15. Fairness through awareness. In Proceedings of the 3rd innovations in theoretical computer science conference, pages 214–226, 2012.
  16. A brief review of domain adaptation. Advances in data science and information engineering: proceedings from ICDATA 2020 and IKE 2020, pages 877–894, 2021.
  17. Causal feature selection for algorithmic fairness. In Proceedings of the 2022 International Conference on Management of Data, pages 276–285, 2022.
  18. Counterfactual fairness in text classification through robustness. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, AIES 2019, Honolulu, HI, USA, January 27-28, 2019, pages 219–226, 2019.
  19. Adversarial learning for counterfactual fairness. Mach. Learn., 112(3):741–763, 2023. doi: 10.1007/S10994-022-06206-8. URL https://doi.org/10.1007/s10994-022-06206-8.
  20. The case for process fairness in learning: Feature selection for fair decision making. In NIPS symposium on machine learning and the law, volume 1, page 11. Barcelona, Spain, 2016.
  21. Equality of opportunity in supervised learning. Advances in neural information processing systems, 29, 2016.
  22. Discretion in hiring. The Quarterly Journal of Economics, 133(2):765–800, 2018.
  23. Nonlinear causal discovery with additive noise models. Advances in neural information processing systems, 21, 2008.
  24. Group-aware threshold adaptation for fair classification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 6988–6995, 2022.
  25. Data preprocessing techniques for classification without discrimination. Knowledge and information systems, 33(1):1–33, 2012.
  26. Fairness in algorithmic decision making: An excursion through the lens of causality. In The World Wide Web Conference, pages 2907–2914, 2019.
  27. Consumer credit-risk models via machine-learning algorithms. Journal of Banking & Finance, 34(11):2767–2787, 2010.
  28. Counterfactual fairness with disentangled causal effect variational autoencoder. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 8128–8136, 2021.
  29. Counterfactual fairness. Advances in neural information processing systems, 30, 2017.
  30. Simfair: A unified framework for fairness-aware multi-label classification. Proceedings of the AAAI Conference on Artificial Intelligence, 37(12):14338–14346, 2023.
  31. Too relaxed to be fair. In International Conference on Machine Learning, pages 6360–6369, 2020.
  32. Survey on causal-based machine learning fairness notions, 2022.
  33. The cost of fairness in binary classification. In Conference on Fairness, accountability and transparency, pages 107–118, 2018.
  34. A penalized likelihood method for balancing accuracy and fairness in predictive policing. In 2018 IEEE international conference on systems, man, and cybernetics (SMC), pages 2454–2459. IEEE, 2018.
  35. Counterfactual identifiability of bijective causal models. arXiv preprint arXiv:2302.02228, 2023.
  36. Causal conceptions of fairness and their consequences. In International Conference on Machine Learning, pages 16848–16887. PMLR, 2022.
  37. Judea Pearl. Causality. Cambridge university press, 2009.
  38. Discrimination-aware data mining. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 560–568, 2008.
  39. Algorithmic Fairness, pages 867–886. 2023.
  40. Causal discovery with continuous additive noise models. 2014.
  41. Post-processing for individual fairness. Advances in Neural Information Processing Systems, 34:25944–25955, 2021.
  42. Causal fairness analysis. arXiv preprint arXiv:2207.11385, 2022.
  43. Counterfactual fairness is basically demographic parity, 2023a.
  44. Counterfactual fairness is basically demographic parity. In Brian Williams, Yiling Chen, and Jennifer Neville, editors, Thirty-Seventh AAAI Conference on Artificial Intelligence, AAAI 2023, pages 14461–14469. AAAI Press, 2023b. doi: 10.1609/AAAI.V37I12.26691. URL https://doi.org/10.1609/aaai.v37i12.26691.
  45. Toward causal representation learning. Proceedings of the IEEE, 109(5):612–634, 2021.
  46. Achieving fairness with a simple ridge penalty, 2021.
  47. A linear non-gaussian acyclic model for causal discovery. Journal of Machine Learning Research, 7(10), 2006.
  48. Counterfactual fairness: removing direct effects through regularization. CoRR, abs/2002.10774, 2020.
  49. Linda F Wightman. Lsac national longitudinal bar passage study. lsac research report series. 1998.
  50. Counterfactual fairness: Unidentification, bound and algorithm. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI-19, pages 1438–1444, 2019.
  51. Fair and optimal classification via post-processing. In International Conference on Machine Learning, pages 37977–38012. PMLR, 2023.
  52. Fairness beyond disparate treatment & disparate impact: Learning classification without disparate mistreatment. In Proceedings of the 26th international conference on world wide web, pages 1171–1180, 2017.
  53. Inherent tradeoffs in learning fair representations, 2022.
  54. Domain generalization: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(4):4396–4415, 2022.
  55. Towards characterizing domain counterfactuals for invertible latent causal models. In The Twelfth International Conference on Learning Representations, 2024. URL https://openreview.net/forum?id=v1VvCWJAL8.
  56. Counterfactually fair representation. Advances in neural information processing systems, 2023.

Summary

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

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