FixCLR: Negative-Class Contrastive Learning for Semi-Supervised Domain Generalization

2025/06/25 by Ha Min Son, Son, Ha Min, Shahbaz Rezaei +3
Computer Science · #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2506.20841

openalex publication_date 2025/06/25 · openalex created_date 2025/10/15 · openalex updated_date 2026/07/28

Abstract

Semi-supervised domain generalization (SSDG) aims to solve the problem of generalizing to out-of-distribution data when only a few labels are available. Due to label scarcity, applying domain generalization methods often underperform. Consequently, existing SSDG methods combine semi-supervised learning methods with various regularization terms. However, these methods do not explicitly regularize to learn domains invariant representations across all domains, which is a key goal for domain generalization. To address this, we introduce FixCLR. Inspired by success in self-supervised learning, we change two crucial components to adapt contrastive learning for explicit domain invariance regularization: utilization of class information from pseudo-labels and using only a repelling term. FixCLR can also be added on top of most existing SSDG and semi-supervised methods for complementary performance improvements. Our research includes extensive experiments that have not been previously explored in SSDG studies. These experiments include benchmarking different improvements to semi-supervised methods, evaluating the performance of pretrained versus non-pretrained models, and testing on datasets with many domains. Overall, FixCLR proves to be an effective SSDG method, especially when combined with other semi-supervised methods.

Citations

• Towards Generalizing to Unseen Domains with Few Labels
• Rethinking the Evaluation Protocol of Domain Generalization
• SoftMatch: Addressing the Quantity-Quality Trade-off in Semi-supervised Learning
• FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning
• Revisiting Pretraining for Semi-Supervised Learning in the Low-Label Regime
• Don't fear the unlabelled: safe semi-supervised learning via simple\n debiasing
• Debiased Learning from Naturally Imbalanced Pseudo-Labels
• FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
• CLDA: Contrastive Learning for Semi-Supervised Domain Adaptation
• Category Contrast for Unsupervised Domain Adaptation in Visual Tasks
• Semi-Supervised Domain Generalization with Stochastic StyleMatch
• SelfReg: Self-supervised Contrastive Regularization for Domain Generalization
• Joint Negative and Positive Learning for Noisy Labels
• Generalizing to Unseen Domains: A Survey on Domain Generalization
• WILDS: A Benchmark of in-the-Wild Distribution Shifts
• A Brief Review of Domain Adaptation
• The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization
• Big Self-Supervised Models are Strong Semi-Supervised Learners
• Bootstrap your own latent: A new approach to self-supervised Learning
• Deep Domain-Adversarial Image Generation for Domain Generalisation
• A Simple Framework for Contrastive Learning of Visual Representations
• FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
• ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring
• NLNL: Negative Learning for Noisy Labels
• MixMatch: A Holistic Approach to Semi-Supervised Learning
• Recognition in Terra Incognita
• Generalizing Across Domains via Cross-Gradient Training
• Deeper, Broader and Artier Domain Generalization
• Deep Hashing Network for Unsupervised Domain Adaptation
• Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization
• Temporal Ensembling for Semi-Supervised Learning
• Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning
• Resnet in Resnet: Generalizing Residual Architectures
• ImageNet Large Scale Visual Recognition Challenge
• ImageNet Large Scale Visual Recognition Challenge

Related

• SelfReg: Self-supervised Contrastive Regularization for Domain Generalization
• Connecting Domains and Contrasting Samples: A Ladder for Domain Generalization
• Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization
• Semi-Supervised Domain Generalization with Stochastic StyleMatch
• Towards Generalizing to Unseen Domains with Few Labels