Add to Chrome
Add to Firefox
Add to EdgeCross Domain Generative Augmentation: Domain Generalization with Latent Diffusion Models
Dec 08, 2023
Despite the huge effort in developing novel regularizers for Domain Generalization (DG), adding simple data augmentation to the vanilla ERM which is a practical implementation of the Vicinal Risk Minimization principle (VRM) \citep{chapelle2000vicinal} outperforms or stays competitive with many of the proposed regularizers. The VRM reduces the estimation error in ERM by replacing the point-wise kernel estimates with a more precise estimation of true data distribution that reduces the gap between data points \textbf{within each domain}. However, in the DG setting, the estimation error of true data distribution by ERM is mainly caused by the distribution shift \textbf{between domains} which cannot be fully addressed by simple data augmentation techniques within each domain. Inspired by this limitation of VRM, we propose a novel data augmentation named Cross Domain Generative Augmentation (CDGA) that replaces the pointwise kernel estimates in ERM with new density estimates in the \textbf{vicinity of domain pairs} so that the gap between domains is further reduced. To this end, CDGA, which is built upon latent diffusion models (LDM), generates synthetic images to fill the gap between all domains and as a result, reduces the non-iidness. We show that CDGA outperforms SOTA DG methods under the Domainbed benchmark. To explain the effectiveness of CDGA, we generate more than 5 Million synthetic images and perform extensive ablation studies including data scaling laws, distribution visualization, domain shift quantification, adversarial robustness, and loss landscape analysis.
