Moment Matching for Multi-Source Domain Adaptation

Conventional unsupervised domain adaptation (UDA) assumes that training data are sampled from a single domain. This neglects the more practical scenario where training data are collected from multiple sources, requiring multi-source domain adaptation. We make three major contributions towards addressing this problem. First, we collect and annotate by far the largest UDA dataset, called DomainNet, which contains six domains and about 0.6 million images distributed among 345 categories, addressing the gap in data availability for multi-source UDA research. Second, we propose a new deep learning approach, Moment Matching for Multi-Source Domain Adaptation M3SDA, which aims to transfer knowledge learned from multiple labeled source domains to an unlabeled target domain by dynamically aligning moments of their feature distributions. Third, we provide new theoretical insights specifically for moment matching approaches in both single and multiple source domain adaptation. Extensive experiments are conducted to demonstrate the power of our new dataset in benchmarking state-of-the-art multi-source domain adaptation methods, as well as the advantage of our proposed model. Dataset and Code are available at \url{http://ai.bu.edu/M3SDA/}.

PDF Abstract ICCV 2019 PDF ICCV 2019 Abstract

Datasets


Introduced in the Paper:

DomainNet

Used in the Paper:

MNIST Office-Home Office-31 Syn2Real
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Multi-Source Unsupervised Domain Adaptation Digits-five M3SDA Accuracy 87.7 # 6
Multi-Source Unsupervised Domain Adaptation Office-31 M3SDA Accuracy 83.7 # 5
Multi-Source Unsupervised Domain Adaptation Office-Caltech10 M3SDA Accuracy 96.4 # 5

Methods


No methods listed for this paper. Add relevant methods here