Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 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

Fairness-enhancing mixed effects deep learning improves fairness on in- and out-of-distribution clustered (non-iid) data (2310.03146v4)

Published 4 Oct 2023 in cs.LG

Abstract: Traditional deep learning (DL) models have two ubiquitous limitations. First, they assume training samples are independent and identically distributed (i.i.d), an assumption often violated in real-world datasets where samples are grouped by shared measurements (e.g., participants or cells). This leads to performance degradation, limited generalization, and covariate confounding, which induces Type 1 and Type 2 errors. Second, DL models typically prioritize overall accuracy, favoring accuracy on the majority, while sacrificing performance for underrepresented subpopulations, leading to unfair, biased models. This is critical to remediate, particularly in models influencing decisions regarding loan approvals and healthcare. To address these issues, we propose the Fair Mixed Effects Deep Learning (Fair MEDL) framework. This framework quantifies cluster-invariant fixed effects (FE) and cluster-specific random effects (RE) through: 1) a cluster adversary for learning invariant FE, 2) a Bayesian neural network for RE, and 3) a mixing function combining FE and RE for final predictions. Fairness is enhanced through the architectural and loss function changes introduced by an adversarial debiasing network. We formally define and demonstrate improved fairness across three metrics on both classification and regression tasks: equalized odds, demographic parity, and counterfactual fairness. Our method also identifies and de-weights confounded covariates, mitigating Type 1 and 2 errors. The framework is comprehensively evaluated across three datasets spanning two industries, including finance and healthcare. The Fair MEDL framework improves fairness by 86.4% for Age, 64.9% for Race, 57.8% for Sex, and 36.2% for Marital status, while maintaining robust predictive performance. Our implementation is publicly available on GitHub.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (24)
  1. The Adult dataset. UCI Machine Learning Repository, 1996. DOI: https://doi.org/10.24432/C5XW20.
  2. Putting fairness principles into practice: Challenges, metrics, and improvements. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, pp.  453–459, 2019.
  3. A snapshot of the frontiers of fairness in machine learning. Communications of the ACM, 63(5):82–89, 2020.
  4. Machine learning in finance, volume 1170. Springer, 2020.
  5. BOHB: Robust and efficient hyperparameter optimization at scale. In International conference on machine learning, pp. 1437–1446. Proceedings of Machine Learning Research, 2018.
  6. Equality of opportunity in supervised learning. In 30th Conference on Neural Information Processing Systems (NIPS 2016), 2016.
  7. The non-iid data quagmire of decentralized machine learning. In International Conference on Machine Learning, pp. 4387–4398. Proceeding of Machine Learning Research, 2020.
  8. Imagenet classification with deep convolutional neural networks. In F. Pereira, C.J. Burges, L. Bottou, and K.Q. Weinberger (eds.), Advances in Neural Information Processing Systems, volume 25. Curran Associates, Inc., 2012.
  9. Fairness and missing values. arXiv preprint arXiv:1905.12728, 2019.
  10. Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review. British Medical Journal, 375, 2021.
  11. Blends: Augmentation of functional magnetic resonance images for machine learning using anatomically constrained warping. Brain Connectivity, 13(2):80–88, 2023a.
  12. Adversarially-regularized mixed effects deep learning (armed) models improve interpretability, performance, and generalization on clustered (non-iid) data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023b.
  13. Balanced gradient training of feed forward networks. Neural Processing Letters, 53:1823–1844, 2021.
  14. Preoperative prediction of lymph node metastasis from clinical DCE MRI of the primary breast tumor using a 4D CNN. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part II 23, pp.  326–334. Springer, 2020.
  15. The book of why: the new science of cause and effect. Penguin (UK), 2018.
  16. A review on fairness in machine learning. ACM Computing Surveys (CSUR), 55(3):1–44, 2022.
  17. IPUMS USA dataset: Version 13.0, 2023. DOI: https://doi.org/10.18128/D010.V13.0.
  18. Machine learning-driven credit risk: a systemic review. Neural Computing and Applications, 34(17):14327–14339, 2022.
  19. Using random effects to account for high-cardinality categorical features and repeated measures in deep neural networks. Advances in Neural Information Processing Systems, 34:25111–25122, 2021.
  20. Attention is all you need. Advances in neural information processing systems, 30, 2017.
  21. Learning non-discriminatory predictors. In Conference on Learning Theory, pp.  1920–1953. Proceedings of Machine Learning Research, 2017.
  22. Mixed Effects Neural Networks (MeNets) With Applications to Gaze Estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2019.
  23. An adversarial training framework for mitigating algorithmic biases in clinical machine learning. NPJ Digital Medicine, 6(1):55, 2023.
  24. Mitigating unwanted biases with adversarial learning. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pp.  335–340, 2018.

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