Source-free Semantic Regularization Learning for Semi-supervised Domain Adaptation

TL;DR

This paper introduces SERL, a novel semi-supervised domain adaptation framework that leverages multiple semantic regularization techniques to improve target domain learning and achieves state-of-the-art results.

Contribution

SERL is the first to integrate probabilistic contrastive regularization, hard-sample mixup, and prediction regularization for effective source-free semi-supervised domain adaptation.

Findings

SERL outperforms existing methods on benchmark datasets.

The combination of three regularization techniques enhances semantic understanding.

SERL achieves state-of-the-art accuracy in SSDA tasks.

Abstract

Semi-supervised domain adaptation (SSDA) has been extensively researched due to its ability to improve classification performance and generalization ability of models by using a small amount of labeled data on the target domain. However, existing methods cannot effectively adapt to the target domain due to difficulty in fully learning rich and complex target semantic information and relationships. In this paper, we propose a novel SSDA learning framework called semantic regularization learning (SERL), which captures the target semantic information from multiple perspectives of regularization learning to achieve adaptive fine-tuning of the source pre-trained model on the target domain. SERL includes three robust semantic regularization techniques. Firstly, semantic probability contrastive regularization (SPCR) helps the model learn more discriminative feature representations from a probabilistic perspective, using semantic information on the target domain to understand the similarities and differences between samples. Additionally, adaptive weights in SPCR can help the model learn the semantic distribution correctly through the probabilities of different samples. To further comprehensively understand the target semantic distribution, we introduce hard-sample mixup regularization (HMR), which uses easy samples as guidance to mine the latent target knowledge contained in hard samples, thereby learning more complete and complex target semantic knowledge. Finally, target prediction regularization (TPR) regularizes the target predictions of the model by maximizing the correlation between the current prediction and the past learned objective, thereby mitigating the misleading of semantic information caused by erroneous pseudo-labels. Extensive experiments on three benchmark datasets demonstrate that our SERL method achieves state-of-the-art performance.