Dynamic Domain Discrepancy Adjustment for Active Multi-Domain Adaptation

TL;DR

This paper introduces D3AAMDA, a novel multi-source domain adaptation method that dynamically adjusts feature alignment and actively selects samples to improve transfer performance with minimal sampling costs.

Contribution

The paper proposes a dynamic modulation mechanism and an active sample selection strategy for multi-source domain adaptation, addressing negative transfer and reducing sampling costs.

Findings

Outperforms existing UDA and ADA methods on standard datasets.

Effectively leverages local source domain features for better alignment.

Reduces sampling costs while maintaining high adaptation accuracy.

Abstract

Multi-source unsupervised domain adaptation (MUDA) aims to transfer knowledge from related source domains to an unlabeled target domain. While recent MUDA methods have shown promising results, most focus on aligning the overall feature distributions across source domains, which can lead to negative effects due to redundant features within each domain. Moreover, there is a significant performance gap between MUDA and supervised methods. To address these challenges, we propose a novel approach called Dynamic Domain Discrepancy Adjustment for Active Multi-Domain Adaptation (D3AAMDA). Firstly, we establish a multi-source dynamic modulation mechanism during the training process based on the degree of distribution differences between source and target domains. This mechanism controls the alignment level of features between each source domain and the target domain, effectively leveraging the local advantageous feature information within the source domains. Additionally, we propose a Multi-source Active Boundary Sample Selection (MABS) strategy, which utilizes a guided dynamic boundary loss to design an efficient query function for selecting important samples. This strategy achieves improved generalization to the target domain with minimal sampling costs. We extensively evaluate our proposed method on commonly used domain adaptation datasets, comparing it against existing UDA and ADA methods. The experimental results unequivocally demonstrate the superiority of our approach.