Training individually fair ML models with Sensitive Subspace Robustness

Abstract

We consider training machine learning models that are fair in the sense that their performance is invariant under certain sensitive perturbations to the inputs. For example, the performance of a resume screening system should be invariant under changes to the gender and/or ethnicity of the applicant. We formalize this notion of algorithmic fairness as a variant of individual fairness and develop a distributionally robust optimization approach to enforce it during training. We also demonstrate the effectiveness of the approach on two ML tasks that are susceptible to gender and racial biases.

ICLR 2020 (spotlight)

References in corpus (17)

• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• Towards Deep Learning Models Resistant to Adversarial Attacks
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• The Measure and Mismeasure of Fairness
• Adversarial Machine Learning at Scale
• Distributional Smoothing with Virtual Adversarial Training
• Certifying Some Distributional Robustness with Principled Adversarial Training
• Robust Wasserstein Profile Inference and Applications to Machine Learning
• Fairness Without Demographics in Repeated Loss Minimization
• Learning Adversarially Fair and Transferable Representations
• Distributionally Robust Logistic Regression
• Censoring Representations with an Adversary
• Minimax Statistical Learning with Wasserstein Distances
• The Limitations of Deep Learning in Adversarial Settings
• Wasserstein Fair Classification
• Variance-based regularization with convex objectives

Cited by in corpus (17)

• Learning Certified Individually Fair Representations
• Exactly Computing the Local Lipschitz Constant of ReLU Networks
• Two Simple Ways to Learn Individual Fairness Metrics from Data
• A Distributionally Robust Approach to Fair Classification
• Toward Operationalizing Pipeline-aware ML Fairness: A Research Agenda for Developing Practical Guidelines and Tools
• SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
• Fairness Evaluation in Text Classification: Machine Learning Practitioner Perspectives of Individual and Group Fairness
• Fair Mixup: Fairness via Interpolation
• Auditing ML Models for Individual Bias and Unfairness
• Does enforcing fairness mitigate biases caused by subpopulation shift?
• Task-Adaptive Neural Network Search with Meta-Contrastive Learning
• Robust Fairness-aware Learning Under Sample Selection Bias
• Fairness with Continuous Optimal Transport
• Individually Fair Ranking
• Fair Training of Decision Tree Classifiers
• Statistical inference for individual fairness
• Individually Fair Gradient Boosting