Fairness Through Robustness: Investigating Robustness Disparity in Deep Learning

Abstract

Deep neural networks (DNNs) are increasingly used in real-world applications (e.g. facial recognition). This has resulted in concerns about the fairness of decisions made by these models. Various notions and measures of fairness have been proposed to ensure that a decision-making system does not disproportionately harm (or benefit) particular subgroups of the population. In this paper, we argue that traditional notions of fairness that are only based on models' outputs are not sufficient when the model is vulnerable to adversarial attacks. We argue that in some cases, it may be easier for an attacker to target a particular subgroup, resulting in a form of \textit{robustness bias}. We show that measuring robustness bias is a challenging task for DNNs and propose two methods to measure this form of bias. We then conduct an empirical study on state-of-the-art neural networks on commonly used real-world datasets such as CIFAR-10, CIFAR-100, Adience, and UTKFace and show that in almost all cases there are subgroups (in some cases based on sensitive attributes like race, gender, etc) which are less robust and are thus at a disadvantage. We argue that this kind of bias arises due to both the data distribution and the highly complex nature of the learned decision boundary in the case of DNNs, thus making mitigation of such biases a non-trivial task. Our results show that robustness bias is an important criterion to consider while auditing real-world systems that rely on DNNs for decision making. Code to reproduce all our results can be found here: \url{https://github.com/nvedant07/Fairness-Through-Robustness}

Accepted at ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2021

References in corpus (20)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• Equality of Opportunity in Supervised Learning
• One weird trick for parallelizing convolutional neural networks
• Counterfactual Fairness
• Certified Adversarial Robustness via Randomized Smoothing
• Fairness of Exposure in Rankings
• Fairness Testing: Testing Software for Discrimination
• Fairness in Machine Learning: Lessons from Political Philosophy
• Towards a Critical Race Methodology in Algorithmic Fairness
• On Fairness and Calibration
• Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers
• Fairness Without Demographics in Repeated Loss Minimization
• Achieving Fairness through Adversarial Learning: an Application to Recidivism Prediction
• From Parity to Preference-based Notions of Fairness in Classification
• Age Progression/Regression by Conditional Adversarial Autoencoder
• On Microtargeting Socially Divisive Ads: A Case Study of Russia-Linked Ad Campaigns on Facebook
• Fairness Under Composition
• Measuring Non-Expert Comprehension of Machine Learning Fairness Metrics
• On the Long-term Impact of Algorithmic Decision Policies: Effort Unfairness and Feature Segregation through Social Learning

Cited by in corpus (5)

• To be Robust or to be Fair: Towards Fairness in Adversarial Training
• Subverting Fair Image Search with Generative Adversarial Perturbations
• UnbiasedNets: A Dataset Diversification Framework for Robustness Bias Alleviation in Neural Networks
• Technical Challenges for Training Fair Neural Networks
• Does Robustness Improve Fairness? Approaching Fairness with Word Substitution Robustness Methods for Text Classification