Learning Cross-domain Generalizable Features by Representation Disentanglement

Abstract

Deep learning models exhibit limited generalizability across different domains. Specifically, transferring knowledge from available entangled domain features(source/target domain) and categorical features to new unseen categorical features in a target domain is an interesting and difficult problem that is rarely discussed in the current literature. This problem is essential for many real-world applications such as improving diagnostic classification or prediction in medical imaging. To address this problem, we propose Mutual-Information-based Disentangled Neural Networks (MIDNet) to extract generalizable features that enable transferring knowledge to unseen categorical features in target domains. The proposed MIDNet is developed as a semi-supervised learning paradigm to alleviate the dependency on labeled data. This is important for practical applications where data annotation requires rare expertise as well as intense time and labor. We demonstrate our method on handwritten digits datasets and a fetal ultrasound dataset for image classification tasks. Experiments show that our method outperforms the state-of-the-art and achieve expected performance with sparsely labeled data.

References in corpus (12)

• Learning Transferable Features with Deep Adaptation Networks
• Deep Domain Confusion: Maximizing for Domain Invariance
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
• Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes
• Adversarial Discriminative Domain Adaptation
• Domain Generalization via Model-Agnostic Learning of Semantic Features
• Understanding disentangling in $\beta$-VAE
• Disentangling factors of variation in deep representations using adversarial training
• Disentangled Representation Learning in Cardiac Image Analysis
• Domain Agnostic Learning with Disentangled Representations
• A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation
• Representation Disentanglement for Multi-task Learning with application to Fetal Ultrasound

Cited by in corpus (4)

• Unsupervised Cross-domain Image Classification by Distance Metric Guided Feature Alignment
• Application of Computer Vision and Machine Learning for Digitized Herbarium Specimens: A Systematic Literature Review
• Decomposing Normal and Abnormal Features of Medical Images into Discrete Latent Codes for Content-Based Image Retrieval
• Embracing the Disharmony in Medical Imaging: A Simple and Effective Framework for Domain Adaptation