基于高斯特征桥的长尾半监督学习几何正则化方法
Geometric Regularization for Long-Tailed Semi-Supervised Learning via Gaussian Feature Bridges
- University of Warwick（华威大学）
- Wuhan University（武汉大学）
- Emory University（埃默里大学）
- Peking University（北京大学）
- Zhejiang University（浙江大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究针对长尾半监督学习的标签分布失衡与伪标签噪声问题，提出高斯桥一致性框架及BridgeMix策略，经基准实验验证可有效提升长尾类性能且具备良好可扩展性。
AI中文摘要：
现实世界中的半监督学习（SSL）常面临长尾标签分布和噪声伪标签的重大挑战，这些问题会阻碍模型泛化并放大确认偏差。本研究提出一种新型框架——高斯桥一致性（GBC），通过在无标签样本与高质量类锚点间构建语义插值路径来应对上述挑战。该方法维护动态原型图谱（Prototype Atlas），存储每类多样且不断演化的有标签和伪标签示例；针对每个无标签样本，GBC在潜在空间中构建类条件高斯特征桥，使学生模型能从不确定预测向可靠类原型遍历平滑轨迹；沿该路径应用桥一致性损失，以强制与几何插值目标分布对齐。此外，还提出BridgeMix这一置信度感知特征混合策略，通过插值样本与锚点对来增强跨样本泛化。在CIFAR10-LT和ImageNet-LT（USB基准）上开展的大量实验验证了GBC在现实长尾SSL设置下的鲁棒性与有效性，其可在不牺牲可扩展性的前提下持续提升长尾类性能。
英文摘要：
Real-world semi-supervised learning (SSL) often encounters significant challenges with long-tailed label distributions and noisy pseudo-labels, which hinder generalization and amplify confirmation bias. In this work, we introduce a novel framework, Gaussian Bridge Consistency (GBC), to address these challenges by constructing semantic interpolation paths between unlabeled samples and high-quality class anchors. Our method maintains a dynamic Prototype Atlas that stores a diverse and evolving set of labeled and pseudo-labeled exemplars per class. For each unlabeled instance, GBC forms a class-conditional Gaussian Feature Bridge in the latent space, enabling the student model to traverse a smooth trajectory from uncertain predictions to reliable class prototypes. A bridge consistency loss is applied along this path to enforce alignment with a geometrically interpolated target distribution. Furthermore, we propose BridgeMix, a confidence-aware feature mixing strategy that interpolates both sample and anchor pairs to amplify cross-sample generalization. Extensive experiments on CIFAR10-LT and ImageNet-LT (USB benchmarks) validate the robustness and effectiveness of GBC under realistic long-tailed SSL settings, consistently improving long tail-class performance without sacrificing scalability.
