用非对称Stein散度提升小样本目标域适应性能
Stein Discrepancy for Unsupervised Domain Adaptation
- 采用仅依赖目标分布得分函数的非对称散度，适配数据稀疏场景
- 在多个基准上，目标数据少时表现优于现有方法
- 支持高斯、GMM、VAE等灵活建模，理论保障收敛与泛化
无监督域适应（UDA）旨在利用相关但未标注的目标域数据提升模型性能。传统方法常通过最小化源域与目标域特征分布间的对称距离（如最大均值差异MMD）实现对齐，但在目标数据稀缺时表现不佳。本文提出一种基于Stein散度的新框架，该散度为非对称度量，仅通过目标分布的得分函数依赖其信息，特别适合低数据目标场景。方法提供核化与对抗两种形式，支持通过高斯、混合高斯或变分自编码器（VAE）灵活建模目标分布。我们推导了目标误差的泛化界及两样本设置下经验Stein散度的收敛速率。实验表明，在多个基准上，本方法在目标数据有限条件下始终优于现有方法。
原文摘要 · Abstract (English)
Unsupervised domain adaptation (UDA) aims to improve model performance on an unlabeled target domain using a related, labeled source domain. A common approach aligns source and target feature distributions by minimizing a distance between them, often using symmetric measures such as maximum mean discrepancy (MMD). However, these methods struggle when target data is scarce. We propose a novel UDA framework that leverages Stein discrepancy, an asymmetric measure that depends on the target distribution only through its score function, making it particularly suitable for low-data target regimes. Our proposed method has kernelized and adversarial forms and supports flexible modeling of the target distribution via Gaussian, GMM, or VAE models. We derive a generalization bound on the target error and a convergence rate for the empirical Stein discrepancy in the two-sample setting. Empirically, our method consistently outperforms prior UDA approaches under limited target data across multiple benchmarks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。