Metric-Learning-Assisted Domain Adaptation

TL;DR

This paper introduces MLA-DA, a novel domain adaptation method that uses a triplet loss with adaptive margin adjustment based on target prediction confidence, leading to improved performance over existing methods.

Contribution

The paper proposes a new metric-learning-assisted domain adaptation approach with an adaptive triplet loss mechanism, addressing limitations of traditional domain alignment assumptions.

Findings

MLA-DA outperforms state-of-the-art methods on four benchmarks.

The adaptive triplet loss improves feature alignment and target risk reduction.

MLA-DA demonstrates stable and robust performance in experiments.

Abstract

Domain alignment (DA) has been widely used in unsupervised domain adaptation. Many existing DA methods assume that a low source risk, together with the alignment of distributions of source and target, means a low target risk. In this paper, we show that this does not always hold. We thus propose a novel metric-learning-assisted domain adaptation (MLA-DA) method, which employs a novel triplet loss for helping better feature alignment. We explore the relationship between the second largest probability of a target sample's prediction and its distance to the decision boundary. Based on the relationship, we propose a novel mechanism to adaptively adjust the margin in the triplet loss according to target predictions. Experimental results show that the use of proposed triplet loss can achieve clearly better results. We also demonstrate the performance improvement of MLA-DA on all four standard benchmarks compared with the state-of-the-art unsupervised domain adaptation methods. Furthermore, MLA-DA shows stable performance in robust experiments.