Domain-Agnostic Feature Modulation for Semi-Supervised Domain Generalization
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913168870080512 |
|---|---|
| author | Amarasinghe, Venuri Bandara, Kalinga Randila, Isun Jayakody, Asini Galappaththige, Chamuditha Jayanga Rodrigo, Ranga |
| author_facet | Amarasinghe, Venuri Bandara, Kalinga Randila, Isun Jayakody, Asini Galappaththige, Chamuditha Jayanga Rodrigo, Ranga |
| contents | Semi-supervised domain generalization (SSDG) leverages a small fraction of labeled data alongside unlabeled data to enhance model generalization. Most of the existing SSDG methods rely on pseudo-labeling (PL) for unlabeled data, often assuming access to domain labels-a privilege not always available. However, domain shifts introduce domain noise, leading to inconsistent PLs that degrade model performance. Methods derived from FixMatch suffer particularly from lower PL accuracy, reducing the effectiveness of unlabeled data. To address this, we tackle the more challenging domain-label agnostic SSDG, where domain labels for unlabeled data are not available during training. First, we propose a feature modulation strategy that enhances class-discriminative features while suppressing domain-specific information. This modulation shifts features toward Similar Average Representations-a modified version of class prototypes-that are robust across domains, encouraging the classifier to distinguish between closely related classes and feature extractor to form tightly clustered, domain-invariant representations. Second, to mitigate domain noise and improve pseudo-label accuracy, we introduce a loss-scaling function that dynamically lowers the fixed confidence threshold for pseudo-labels, optimizing the use of unlabeled data. With these key innovations, our approach achieves significant improvements on four major domain generalization benchmarks-even without domain labels. We will make the code available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_20897 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Domain-Agnostic Feature Modulation for Semi-Supervised Domain Generalization Amarasinghe, Venuri Bandara, Kalinga Randila, Isun Jayakody, Asini Galappaththige, Chamuditha Jayanga Rodrigo, Ranga Computer Vision and Pattern Recognition Semi-supervised domain generalization (SSDG) leverages a small fraction of labeled data alongside unlabeled data to enhance model generalization. Most of the existing SSDG methods rely on pseudo-labeling (PL) for unlabeled data, often assuming access to domain labels-a privilege not always available. However, domain shifts introduce domain noise, leading to inconsistent PLs that degrade model performance. Methods derived from FixMatch suffer particularly from lower PL accuracy, reducing the effectiveness of unlabeled data. To address this, we tackle the more challenging domain-label agnostic SSDG, where domain labels for unlabeled data are not available during training. First, we propose a feature modulation strategy that enhances class-discriminative features while suppressing domain-specific information. This modulation shifts features toward Similar Average Representations-a modified version of class prototypes-that are robust across domains, encouraging the classifier to distinguish between closely related classes and feature extractor to form tightly clustered, domain-invariant representations. Second, to mitigate domain noise and improve pseudo-label accuracy, we introduce a loss-scaling function that dynamically lowers the fixed confidence threshold for pseudo-labels, optimizing the use of unlabeled data. With these key innovations, our approach achieves significant improvements on four major domain generalization benchmarks-even without domain labels. We will make the code available. |
| title | Domain-Agnostic Feature Modulation for Semi-Supervised Domain Generalization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.20897 |