潜藏何物？利用扩散潜空间实现领域泛化

What's in a Latent? Leveraging Diffusion Latent Space for Domain Generalization

March 9, 2025

作者: Xavier Thomas, Deepti Ghadiyaram

cs.AI

摘要

领域泛化旨在开发能够适应新颖且未见过的数据分布的模型。在本研究中，我们探讨了模型架构与预训练目标如何影响特征的丰富性，并提出了一种有效利用这些特征进行领域泛化的方法。具体而言，给定一个预训练的特征空间，我们首先以无监督的方式发现捕捉领域特定变化的潜在领域结构，称之为伪域。接着，我们通过将这些互补的伪域表示融入现有分类器，使其更易于适应多样化的未见测试域。我们分析了不同预训练特征空间在捕获领域特定差异方面的差异。实证研究表明，在缺乏明确领域标签的情况下，扩散模型提取的特征在区分领域方面表现卓越，并能捕捉细微的领域特定信息。在五个数据集上，我们展示了这一极其简单的框架相较于标准基线经验风险最小化（ERM），在未见域上的泛化能力提升显著，最高测试准确率提升超过4%。尤为重要的是，我们的方法在训练过程中无需访问领域标签，却超越了大多数依赖领域标签的算法。

English

Domain Generalization aims to develop models that can generalize to novel and unseen data distributions. In this work, we study how model architectures and pre-training objectives impact feature richness and propose a method to effectively leverage them for domain generalization. Specifically, given a pre-trained feature space, we first discover latent domain structures, referred to as pseudo-domains, that capture domain-specific variations in an unsupervised manner. Next, we augment existing classifiers with these complementary pseudo-domain representations making them more amenable to diverse unseen test domains. We analyze how different pre-training feature spaces differ in the domain-specific variances they capture. Our empirical studies reveal that features from diffusion models excel at separating domains in the absence of explicit domain labels and capture nuanced domain-specific information. On 5 datasets, we show that our very simple framework improves generalization to unseen domains by a maximum test accuracy improvement of over 4% compared to the standard baseline Empirical Risk Minimization (ERM). Crucially, our method outperforms most algorithms that access domain labels during training.