革新知能統合研究センター

Speaker: Feng Liu, University of Technology Sydney/Imperfect Information Learning Team, RIKEN AIP

Title: Butterfly: A Panacea for All Difficulties in Wildly Unsupervised Domain Adaptation

Abs: In unsupervised domain adaptation (UDA), classifiers for the target domain (TD) are trained with clean labeled data from the source domain (SD) and unlabeled data from TD. However, in the wild, it is hard to acquire a large amount of perfectly clean labeled data in SD given limited budget. Hence, we consider a new, more realistic and more challenging problem setting, where classifiers have to be trained with noisy labeled data from SD and unlabeled data from TD—we name it wildly UDA (WUDA). We show that WUDA provably ruins all UDA methods if taking no care of label noise in SD, and to this end, we propose a Butterfly framework, a panacea for all difficulties in WUDA. Butterfly maintains four models (e.g., deep networks) simultaneously, where two take care of all adaptations (i.e., noisy-to-clean, labeled-to-unlabeled, and SD-to-TD-distributional) and then the other two can focus on classification in TD. As a consequence, Butterfly possesses all the necessary components for all the challenges in WUDA. Experiments demonstrate that under WUDA, Butterfly significantly outperforms existing baseline methods.

Bio: Feng Liu received his M.Sc. degree in probability and statistics and a B.Sc. degree in pure mathematics from Lanzhou University, China, in 2015 and 2013, respectively. He is working toward a PhD degree with the University of Technology Sydney, Australia. His research interests include domain adaptation and two-sample test. He is a Member of Centre for Artificial Intelligence (CAI) in UTS.