A Multi-View Consistency Framework with Semi-Supervised Domain Adaptation

TL;DR

This paper proposes a multi-view consistency framework with semi-supervised domain adaptation that improves classification accuracy across domains by leveraging data augmentation, debiasing, pseudo-negative labels, and cross-domain feature alignment.

Contribution

It introduces a novel multi-view consistency approach combining debiasing, pseudo-negative labels, and cross-domain affinity learning for semi-supervised domain adaptation.

Findings

Outperforms existing methods on DomainNet and Office-Home datasets.

Enhances model performance by aligning features across domains.

Reduces annotation costs and improves adaptability in industrial applications.

Abstract

Semi-Supervised Domain Adaptation (SSDA) leverages knowledge from a fully labeled source domain to classify data in a partially labeled target domain. Due to the limited number of labeled samples in the target domain, there can be intrinsic similarity of classes in the feature space, which may result in biased predictions, even when the model is trained on a balanced dataset. To overcome this limitation, we introduce a multi-view consistency framework, which includes two views for training strongly augmented data. One is a debiasing strategy for correcting class-wise prediction probabilities according to the prediction performance of the model. The other involves leveraging pseudo-negative labels derived from the model predictions. Furthermore, we introduce a cross-domain affinity learning aimed at aligning features of the same class across different domains, thereby enhancing overall performance. Experimental results demonstrate that our method outperforms the competing methods on two standard domain adaptation datasets, DomainNet and Office-Home. Combining unsupervised domain adaptation and semi-supervised learning offers indispensable contributions to the industrial sector by enhancing model adaptability, reducing annotation costs, and improving performance.