Towards Assumption-free Bias Mitigation
Not on the map. Why?
Abstract (arXiv)
Despite the impressive prediction ability, machine learning models show discrimination towards certain demographics and suffer from unfair prediction behaviors. To alleviate the discrimination, extensive studies focus on eliminating the unequal distribution of sensitive attributes via multiple approaches. However, due to privacy concerns, sensitive attributes are often either unavailable or missing in real-world scenarios. Therefore, several existing works alleviate the bias without sensitive attributes. Those studies face challenges, either in inaccurate predictions of sensitive attributes or the need to mitigate unequal distribution of manually defined non-sensitive attributes related to bias. The latter requires strong assumptions about the correlation between sensitive and non-sensitive attributes. As data distribution and task goals vary, the strong assumption on non-sensitive attributes may not be valid and require domain expertise. In this work, we propose an assumption-free framework to detect the related attributes automatically by modeling feature interaction for bias mitigation. The proposed framework aims to mitigate the unfair impact of identified biased feature interactions. Experimental results on four real-world datasets demonstrate that our proposed framework can significantly alleviate unfair prediction behaviors by considering biased feature interactions.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 44 of its 72 references that are in the map. First posted on arXiv 9 Jul 2023.
Builds on
44 of 72 references.
References · 44
- 2017Attention Is All You Needcited by 60,361
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
- 2017Counterfactual Fairnesscited by 804
- 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
- 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 523
- 2019Composition-based Multi-Relational Graph Convolutional Networkscited by 396
- 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 394
- 2017Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairnesscited by 384
- 2018Fairness Without Demographics in Repeated Loss Minimizationcited by 364
- 2018Learning Adversarially Fair and Transferable Representationscited by 353
32 more
- 2018AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networkscited by 320
- 2020Fairness without Demographics through Adversarially Reweighted Learningcited by 251
- 2015Censoring Representations with an Adversarycited by 192
- 2018Adversarial Removal of Demographic Attributes from Text Datacited by 170
- 2019Fair Regression: Quantitative Definitions and Reduction-based Algorithmscited by 168
- 2018FairGAN: Fairness-aware Generative Adversarial Networkscited by 164
- 2021A survey on datasets for fairness-aware machine learningcited by 157
- 2018Counterfactual Fairness in Text Classification through Robustnesscited by 151
- 2017A Convex Framework for Fair Regressioncited by 148
- 2019On the Fairness of Disentangled Representationscited by 144
- 2019Wasserstein Fair Classificationcited by 131
- 2019Fairness in Recommendation Ranking through Pairwise Comparisonscited by 127
- 2019Compositional Fairness Constraints for Graph Embeddingscited by 126
- 2019Fairness in Deep Learning: A Computational Perspectivecited by 123
- 2019Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognitioncited by 117
- 2019FiBiNET: Combining Feature Importance and Bilinear feature Interaction for Click-Through Rate Predictioncited by 83
- 2017Detecting Statistical Interactions from Neural Network Weightscited by 80
- 2018Achieving Fairness through Adversarial Learning: an Application to Recidivism Predictioncited by 77
- 2019Learning Optimal and Fair Decision Trees for Non-Discriminative Decision-Makingcited by 75
- 2021FairFil: Contrastive Neural Debiasing Method for Pretrained Text Encoderscited by 75
- 2019Putting Fairness Principles into Practice: Challenges, Metrics, and Improvementscited by 70
- 2021Fairness via Representation Neutralizationcited by 63
- 2020Counterfactual Fairness with Disentangled Causal Effect Variational Autoencodercited by 45
- 2021Towards Fair Classifiers Without Sensitive Attributes: Exploring Biases in Related Featurescited by 45
- 2020Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systemscited by 38
- 2020Feature Interaction Interpretability: A Case for Explaining Ad-Recommendation Systems via Neural Interaction Detectioncited by 32
- 2019Paper Matching with Local Fairness Constraintscited by 30
- 2016Disciplined Convex-Concave Programmingcited by 29
- 2021Modeling Techniques for Machine Learning Fairness: A Surveycited by 17
- 2020Counterfactual fairness: removing direct effects through regularizationcited by 14
- 2022Mitigating Relational Bias on Knowledge Graphscited by 6
- 2020Skewness Ranking Optimization for Personalized Recommendationcited by 4
Built on by
Citing papers per year: 2023: 1.
1 paper.