Adversarial Unsupervised Domain Adaptation with Conditional and Label Shift: Infer, Align and Iterate

Xiaofeng Liu, Zhenhua Guo, Site Li, Fangxu Xing, Jia You, C-C Jay Kuo, Georges El Fakhri, Jonghye Woo

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

In this work, we propose an adversarial unsupervised domain adaptation (UDA) method under inherent conditional and label shifts, in which we aim to align the distributions w.r.t. both p(x|y) and p(y). Since labels are inaccessible in a target domain, conventional adversarial UDA methods assume that p(y) is invariant across domains and rely on aligning p(x) as an alternative to the p(x|y) alignment. To address this, we provide a thorough theoretical and empirical analysis of the conventional adversarial UDA methods under both conditional and label shifts, and propose a novel and practical alternative optimization scheme for adversarial UDA. Specifically, we infer the marginal p(y) and align p(x|y) iteratively at the training stage, and precisely align the posterior p(y|x) at the testing stage. Our experimental results demonstrate its effectiveness on both classification and segmentation UDA and partial UDA.
Original languageEnglish
Title of host publicationProc. of ICCV'2021
PublisherIEEE
Pages10347-10356
DOIs
Publication statusPublished - Oct 2021
Event2021 IEEE/CVF Int. Conf. on Computer Vision - Montreal, QC, Canada
Duration: 10 Oct 202117 Oct 2021

Conference

Conference2021 IEEE/CVF Int. Conf. on Computer Vision
Abbreviated titleICCV
Country/TerritoryCanada
Period10/10/2117/10/21

Fingerprint

Dive into the research topics of 'Adversarial Unsupervised Domain Adaptation with Conditional and Label Shift: Infer, Align and Iterate'. Together they form a unique fingerprint.

Cite this