InclusiveFaceNet: Improving Face Attribute Detection with Race and Gender Diversity

Abstract

We demonstrate an approach to face attribute detection that retains or improves attribute detection accuracy across gender and race subgroups by learning demographic information prior to learning the attribute detection task. The system, which we call InclusiveFaceNet, detects face attributes by transferring race and gender representations learned from a held-out dataset of public race and gender identities. Leveraging learned demographic representations while withholding demographic inference from the downstream face attribute detection task preserves potential users' demographic privacy while resulting in some of the best reported numbers to date on attribute detection in the Faces of the World and CelebA datasets.

Presented as a talk at the 2018 Workshop on Fairness, Accountability, and Transparency in Machine Learning (FAT/ML 2018)

References in corpus (2)

• Equality of Opportunity in Supervised Learning
• Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations

Cited by in corpus (24)

• 50 Years of Test (Un)fairness: Lessons for Machine Learning
• FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age
• Women also Snowboard: Overcoming Bias in Captioning Models
• A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle
• Understanding Unequal Gender Classification Accuracy from Face Images
• Deep Learning for Face Recognition: Pride or Prejudiced?
• Robustness Disparities in Commercial Face Detection
• Improving the Fairness of Deep Generative Models without Retraining
• Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation
• Identifying Bias in AI using Simulation
• Fair Generative Modeling via Weak Supervision
• Demographic Bias: A Challenge for Fingervein Recognition Systems?
• Joint Fairness Model with Applications to Risk Predictions for Under-represented Populations
• Contrastive Examples for Addressing the Tyranny of the Majority
• Biases in Data Science Lifecycle
• Representation Matters: Assessing the Importance of Subgroup Allocations in Training Data
• Comparing Human and Machine Bias in Face Recognition
• Technical Challenges for Training Fair Neural Networks
• On the effect of age perception biases for real age regression
• Inclusive GAN: Improving Data and Minority Coverage in Generative Models
• Fairness in Missing Data Imputation
• Fairness in Deep Learning: A Computational Perspective
• Fairness-aware Federated Minimax Optimization with Convergence Guarantee
• Harnessing Geometric Constraints from Emotion Labels to improve Face Verification