Unsupervised domain adaptation (UDA) is a challenging problem in machine learning where the model is trained on a source domain with labeled data and tested on a target domain with unlabeled data. In recent years, UDA has received significant attention from the research community due to its applicability in various real-world scenarios. This book provides a comprehensive review of state-of-the-art UDA methods and explores new variants of UDA that have the potential to advance the field.
The book begins with a clear introduction to the UDA problem and is mainly organized into four technical sections, each focused on a specific piece of UDA research. The first section covers criterion optimization-based UDA, which aims to learn domain-invariant representations by minimizing the discrepancy between source and target domains. The second section discusses bi-classifier adversarial learning-based UDA, which creatively leverages adversarial learning by conducting a minimax game between the feature extractor and two task classifiers. The third section introduces source-free UDA, a novel UDA setting that does not require any raw data from the source domain. The fourth section presents active learning for UDA, which combines domain adaptation and active learning to reduce the amount of labeled data needed for adaptation.
This book is suitable for researchers, graduate students, and practitioners who are interested in UDA and its applications in various fields, primarily in computer vision. The chapters are authored by leading experts in the field and provide a comprehensive and in-depth analysis of the current UDA methods and new directions for future research. With its broad coverage and cutting-edge research, this book is a valuable resource for anyone looking to advance their knowledge of UDA.
Jingjing Li is currently a professor with the School of Computer Science and Engineering, University of Electronic Science and Technology of China (UESTC). He received his B.Eng., M.Sc. and Ph.D. degrees from UESTC in 2010, 2013, and 2017, respectively. His research interests are in the area of domain adaptation and zero-shot learning. He has co/authored more than 70 peer-reviewed papers, such as IEEE TPAMI, IEEE TIP, IEEE TKDE, CVPR, ICCV, AAAI, IJCAI, and ACM Multimedia. He won Excellent Doctoral Dissertation Award of Chinese Institute of Electronics in 2018.
Lei Zhu is currently a professor with the School of Electronic and Information Engineering, Tongji University. He received his B.Eng. and Ph.D. degrees from Wuhan University of Technology in 2009 and Huazhong University Science and Technology in 2015, respectively. He was a Research Fellow at the University of Queensland (2016–2017). His research interests are in the area of large-scale multimedia contentanalysis and retrieval. Zhu has co/authored more than 100 peer-reviewed papers, such as ACM SIGIR, ACM MM, IEEE TPAMI, IEEE TIP, IEEE TKDE, and ACM TOIS. His publications have attracted more than 5,600 Google citations. At present, he serves as the Associate Editor of IEEE TBD, ACM TOMM, and Information Sciences. He has served as the Area Chair, Senior Program Committee or reviewer for more than 40 well-known international journals and conferences. He won ACM SIGIR 2019 Best Paper Honorable Mention Award, ADMA 2020 Best Paper Award, ChinaMM 2022 Best Student Paper Award, ACM China SIGMM Rising Star Award, Shandong Provincial Entrepreneurship Award for Returned Students, and Shandong Provincial AI Outstanding Youth Award.
Zhekai Du is currently a third-year Ph.D. student with the School of Computer Science and Engineering, University of Electronic Science and Technology of China (UESTC). His research interests are domain adaptation, domain generalization, and their applications in computer vision. He received his B.Eng. degree from UESTC in 2018. He has co/authored dozens of papers at the top conferences and journals, like CVPR, ACM Multimedia, ECCV, AAAI, and IEEE TPAMI.
