Individual Fairness in Bayesian Neural Networks
Not on the map. Why?
Abstract (arXiv)
We study Individual Fairness (IF) for Bayesian neural networks (BNNs). Specifically, we consider the ε-δ-individual fairness notion, which requires that, for any pair of input points that are ε-similar according to a given similarity metrics, the output of the BNN is within a given tolerance δ>0. We leverage bounds on statistical sampling over the input space and the relationship between adversarial robustness and individual fairness to derive a framework for the systematic estimation of ε-δ-IF, designing Fair-FGSM and Fair-PGD as global,fairness-aware extensions to gradient-based attacks for BNNs. We empirically study IF of a variety of approximately inferred BNNs with different architectures on fairness benchmarks, and compare against deterministic models learnt using frequentist techniques. Interestingly, we find that BNNs trained by means of approximate Bayesian inference consistently tend to be markedly more individually fair than their deterministic counterparts.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 20 of its 28 references that are in the map. First posted on arXiv 21 Apr 2023.
Builds on
20 of 28 references.
References · 20
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
- 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
- 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
- 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 443
- 2018Fast and Scalable Bayesian Deep Learning by Weight-Perturbation in Adamcited by 131
- 2019Training individually fair ML models with Sensitive Subspace Robustnesscited by 73
- 2020Learning Certified Individually Fair Representationscited by 69
- 2018Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Networkcited by 68
- 2019Metric Learning for Individual Fairnesscited by 51
8 more
- 2020Verifying Individual Fairness in Machine Learning Modelscited by 49
- 2020Robustness of Bayesian Neural Networks to Gradient-Based Attackscited by 48
- 2017MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examplescited by 46
- 2020Probabilistic Safety for Bayesian Neural Networkscited by 41
- 2019Statistical Guarantees for the Robustness of Bayesian Neural Networkscited by 35
- 2021Bayesian Inference with Certifiable Adversarial Robustnesscited by 26
- 2022Individual Fairness Guarantees for Neural Networkscited by 23
- 2020Individual Fairness Revisited: Transferring Techniques from Adversarial Robustnesscited by 17
Built on by
Citing papers per year: 2023: 3.
3 papers.