Unsupervised multi-task domain adaptation

Shih Min Yang, Mei Chen Yeh

研究成果: 書貢獻/報告類型會議論文篇章

摘要

With abundant labeled data, deep convolutional neural networks have shown great success in various image recognition tasks. However, these models are often less powerful when applied to novel datasets due to a phenomenon known as domain shift. Unsupervised domain adaptation methods aim to address this problem, allowing deep models trained on the labeled source domain to be used on a different target domain (without labels). In this paper, we investigate whether the generalization ability of an unsupervised domain adaptation method can be improved through multi-task learning, with learned features required to be both domain invariant and discriminative for multiple different but relevant tasks. Experiments evaluating two fundamental recognition tasks-image recognition and segmentation-show that the generalization ability empowered by multi-task learning may not benefit recognition when the model is directly applied on the target domain, but the multi-task learning setting can boost the performance of state-of-the-art unsupervised domain adaptation methods by a non-negligible margin.

原文英語
主出版物標題Proceedings of ICPR 2020 - 25th International Conference on Pattern Recognition
發行者Institute of Electrical and Electronics Engineers Inc.
頁面1679-1685
頁數7
ISBN(電子)9781728188089
DOIs
出版狀態已發佈 - 2020
事件25th International Conference on Pattern Recognition, ICPR 2020 - Virtual, Milan, 意大利
持續時間: 2021 一月 102021 一月 15

出版系列

名字Proceedings - International Conference on Pattern Recognition
ISSN(列印)1051-4651

會議

會議25th International Conference on Pattern Recognition, ICPR 2020
國家/地區意大利
城市Virtual, Milan
期間2021/01/102021/01/15

ASJC Scopus subject areas

  • 電腦視覺和模式識別

指紋

深入研究「Unsupervised multi-task domain adaptation」主題。共同形成了獨特的指紋。

引用此