Conditional Bures Metric for Domain Adaptation

2021/07/31 by You-Wei Luo, Luo, You-Wei, Chuan-Xian Ren +1 · 4 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Cancer-related molecular mechanisms research #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and ELM

paper · pdf · doi:10.48550/arxiv.2108.00302

openalex publication_date 2021/07/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01

Abstract

As a vital problem in classification-oriented transfer, unsupervised domain adaptation (UDA) has attracted widespread attention in recent years. Previous UDA methods assume the marginal distributions of different domains are shifted while ignoring the discriminant information in the label distributions. This leads to classification performance degeneration in real applications. In this work, we focus on the conditional distribution shift problem which is of great concern to current conditional invariant models. We aim to seek a kernel covariance embedding for conditional distribution which remains yet unexplored. Theoretically, we propose the Conditional Kernel Bures (CKB) metric for characterizing conditional distribution discrepancy, and derive an empirical estimation for the CKB metric without introducing the implicit kernel feature map. It provides an interpretable approach to understand the knowledge transfer mechanism. The established consistency theory of the empirical estimation provides a theoretical guarantee for convergence. A conditional distribution matching network is proposed to learn the conditional invariant and discriminative features for UDA. Extensive experiments and analysis show the superiority of our proposed model.

Citations

• Learning Transferable Features with Deep Adaptation Networks
• ImageNet classification with deep convolutional neural networks
• A Survey on Transfer Learning
• Conditional Adversarial Domain Adaptation
• On Learning Invariant Representation for Domain Adaptation

Cited by

• Partial Domain Adaptation via Importance Sampling-based Shift Correction
• A Generalized Label Shift Perspective for Cross-Domain Gaze Estimation
• Bi-level Unbalanced Optimal Transport for Partial Domain Adaptation
• OT Score: An OT based Confidence Score for Prototype-Assisted Source Free Unsupervised Domain Adaptation

Related

• COD: Learning Conditional Invariant Representation for Domain Adaptation Regression
• Mapping conditional distributions for domain adaptation under generalized target shift
• Maximizing Conditional Independence for Unsupervised Domain Adaptation
• CASUAL: Conditional Support Alignment for Domain Adaptation with Label Shift
• Domain Adaptation and Image Classification via Deep Conditional Adaptation Network