mlsanityData to 14 Oct 2026

Fairness without Demographics through Learning Graph of Gradients

arXiv 2412.03706 ↗First posted 4 Dec 2024cs.LG stat.MLcited by 4

Not on the map. Why?

Abstract (arXiv)

Machine learning systems are notoriously prone to biased predictions about certain demographic groups, leading to algorithmic fairness issues. Due to privacy concerns and data quality problems, some demographic information may not be available in the training data and the complex interaction of different demographics can lead to a lot of unknown minority subpopulations, which all limit the applicability of group fairness. Many existing works on fairness without demographics assume the correlation between groups and features. However, we argue that the model gradients are also valuable for fairness without demographics. In this paper, we show that the correlation between gradients and groups can help identify and improve group fairness. With an adversarial weighting architecture, we construct a graph where samples with similar gradients are connected and learn the weights of different samples from it. Unlike the surrogate grouping methods that cluster groups from features and labels as proxy sensitive attribute, our method leverages the graph structure as a soft grouping mechanism, which is much more robust to noises. The results show that our method is robust to noise and can improve fairness significantly without decreasing the overall accuracy too much.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 19 of its 38 references that are in the map. First posted on arXiv 4 Dec 2024.

Builds on

19 of 38 references.

References · 19

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  2. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 672
  3. 2018Fairness Without Demographics in Repeated Loss Minimizationcited by 364
  4. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  5. 2020Environment Inference for Invariant Learningcited by 306
  6. 2020Fairness without Demographics through Adversarially Reweighted Learningcited by 251
  7. 2018Why Is My Classifier Discriminatory?cited by 143
  8. 2020FairBatch: Batch Selection for Model Fairnesscited by 99
  9. 2021EDITS: Modeling and Mitigating Data Bias for Graph Neural Networkscited by 92
  10. 2021Fairness via Representation Neutralizationcited by 63
  11. 2022Improving Fairness in Graph Neural Networks via Mitigating Sensitive Attribute Leakagecited by 61
  12. 2022Learning Debiased Classifier with Biased Committeecited by 51
7 more
  1. 2019AdaFair: Cumulative Fairness Adaptive Boostingcited by 45
  2. 2021Towards Fair Classifiers Without Sensitive Attributes: Exploring Biases in Related Featurescited by 45
  3. 2022On Learning Fairness and Accuracy on Multiple Subgroupscited by 32
  4. 2021Fair Classification with Adversarial Perturbationscited by 23
  5. 2021Fairness without the sensitive attribute via Causal Variational Autoencodercited by 21
  6. 2021Group-Aware Threshold Adaptation for Fair Classificationcited by 21
  7. 2022Interpreting Unfairness in Graph Neural Networks via Training Node Attributioncited by 18

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2025Fairness in Augmented Graph Learning: A Surveycited by 2
  2. 2025Benchmarking Fairness-aware Graph Neural Networks in Knowledge Graphscited by 2
  3. 2026Ethical Fairness in Ubiquitous Health Sensing without Known Attributescited by 1
  4. 2026Grad2Fair: A Gradient-driven Approach for Graph Fairness without Demographicscited by 0