Target Consistency for Domain Adaptation: when Robustness meets Transferability

2020-06-25 · Yassine Ouali, Victor Bouvier, Myriam Tami, Céline Hudelot

Learning Invariant Representations has been successfully applied for reconciling a source and a target domain for Unsupervised Domain Adaptation. By investigating the robustness of such methods under the prism of the cluster assumption, we bring new evidence that invariance with a low source risk does not guarantee a well-performing target classifier. More precisely, we show that the cluster assumption is violated in the target domain despite being maintained in the source domain, indicating a lack of robustness of the target classifier. To address this problem, we demonstrate the importance of enforcing the cluster assumption in the target domain, named Target Consistency (TC), especially when paired with Class-Level InVariance (CLIV). Our new approach results in a significant improvement, on both image classification and segmentation benchmarks, over state-of-the-art methods based on invariant representations. Importantly, our method is flexible and easy to implement, making it a complementary technique to existing approaches for improving transferability of representations.

📄 PDF Abstract BibTeX arXiv:2006.14263

Code (0)

등록된 구현이 없습니다.

Tasks

Domain Adaptation image-classification Image Classification Unsupervised Domain Adaptation

Preserving Semantic Consistency in Unsupervised Domain Adaptation Using Generative Adversarial Networks

2021-04-28 · Mohammad Mahfujur Rahman, Clinton Fookes, Sridha Sridharan

Unsupervised domain adaptation seeks to mitigate the distribution discrepancy between source and target domains, given labeled samples of the source domain and unlabeled samples of the target domain. Generative adversari…

Domain Adaptation Generative Adversarial Network Unsupervised Domain Adaptation

DRIVE: Dual-Robustness via Information Variability and Entropic Consistency in Source-Free Unsupervised Domain Adaptation

2024-11-24 · Ruiqiang Xiao, Songning Lai, Yijun Yang, Jiemin Wu 외

Adapting machine learning models to new domains without labeled data, especially when source data is inaccessible, is a critical challenge in applications like medical imaging, autonomous driving, and remote sensing. Thi…

Autonomous Driving Domain Adaptation Unsupervised Domain Adaptation

Multi Task Consistency Guided Source-Free Test-Time Domain Adaptation Medical Image Segmentation

2023-10-18 · Yanyu Ye, Zhenxi Zhang, Wei Wei, Chunna Tian

Source-free test-time adaptation for medical image segmentation aims to enhance the adaptability of segmentation models to diverse and previously unseen test sets of the target domain, which contributes to the generaliza…

Domain Adaptation Image Segmentation Medical Image Segmentation Segmentation+2

TWINs: Two Weighted Inconsistency-reduced Networks for Partial Domain Adaptation

2018-12-18 · Toshihiko Matsuura, Kuniaki Saito, Tatsuya Harada

The task of unsupervised domain adaptation is proposed to transfer the knowledge of a label-rich domain (source domain) to a label-scarce domain (target domain). Matching feature distributions between different domains i…

Domain Adaptation General Classification Partial Domain Adaptation Unsupervised Domain Adaptation+1

Surprisingly Simple Semi-Supervised Domain Adaptation with Pretraining and Consistency

2021-01-29 · Samarth Mishra, Kate Saenko, Venkatesh Saligrama

Most modern unsupervised domain adaptation (UDA) approaches are rooted in domain alignment, i.e., learning to align source and target features to learn a target domain classifier using source labels. In semi-supervised d…

Domain Adaptation Semi-supervised Domain Adaptation Unsupervised Domain Adaptation