Additive Adversarial Learning for Unbiased Authentication
arXiv:1905.06517
Abstract
Authentication is a task aiming to confirm the truth between data instances and personal identities. Typical authentication applications include face recognition, person re-identification, authentication based on mobile devices and so on. The recently-emerging data-driven authentication process may encounter undesired biases, i.e., the models are often trained in one domain (e.g., for people wearing spring outfits) while required to apply in other domains (e.g., they change the clothes to summer outfits). To address this issue, we propose a novel two-stage method that disentangles the class/identity from domain-differences, and we consider multiple types of domain-difference. In the first stage, we learn disentangled representations by a one-versus-rest disentangle learning (OVRDL) mechanism. In the second stage, we improve the disentanglement by an additive adversarial learning (AAL) mechanism. Moreover, we discuss the necessity to avoid a learning dilemma due to disentangling causally related types of domain-difference. Comprehensive evaluation results demonstrate the effectiveness and superiority of the proposed method.
IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'2019)
References in corpus (16)
- Deep Visual Domain Adaptation: A Survey
- Equality of Opportunity in Supervised Learning
- Person Re-identification: Past, Present and Future
- The relativistic discriminator: a key element missing from standard GAN
- Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
- Domain Adaptation for Visual Applications: A Comprehensive Survey
- The Frontiers of Fairness in Machine Learning
- Learning Adversarially Fair and Transferable Representations
- Achieving Fairness through Adversarial Learning: an Application to Recidivism Prediction
- CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
- Zero-Shot Learning by Generating Pseudo Feature Representations
- FairGAN: Fairness-aware Generative Adversarial Networks
- Multi-target Unsupervised Domain Adaptation without Exactly Shared Categories
- DiDA: Disentangled Synthesis for Domain Adaptation
- Transfer Metric Learning: Algorithms, Applications and Outlooks
- Network Inference from Temporal-Dependent Grouped Observations