Teacher-Student Competition for Unsupervised Domain Adaptation

TL;DR

This paper introduces a Teacher-Student Competition framework for unsupervised domain adaptation, where a student learns target-specific features through a competitive pseudo-labeling process with a teacher, improving adaptation performance.

Contribution

The paper proposes a novel TSC approach that employs a competition mechanism between teacher and student networks for better pseudo-label selection in UDA.

Findings

Significantly outperforms state-of-the-art methods on Office-31.

Achieves superior results on ImageCLEF-DA.

Demonstrates effectiveness of competition mechanism in pseudo-labeling.

Abstract

With the supervision from source domain only in class-level, existing unsupervised domain adaptation (UDA) methods mainly learn the domain-invariant representations from a shared feature extractor, which causes the source-bias problem. This paper proposes an unsupervised domain adaptation approach with Teacher-Student Competition (TSC). In particular, a student network is introduced to learn the target-specific feature space, and we design a novel competition mechanism to select more credible pseudo-labels for the training of student network. We introduce a teacher network with the structure of existing conventional UDA method, and both teacher and student networks compete to provide target pseudo-labels to constrain every target sample's training in student network. Extensive experiments demonstrate that our proposed TSC framework significantly outperforms the state-of-the-art domain adaptation methods on Office-31 and ImageCLEF-DA benchmarks.