To sample from an unconditionally trained Denoising Diffusion Probabilistic Model (DDPM), classifier guidance adds conditional information during sampling, but the gradients from classifiers, especially those not trained on noisy images, are often unstable. This study conducts a gradient analysis comparing robust and non-robust classifiers, as well as multiple gradient stabilization techniques. Experimental results demonstrate that these techniques significantly improve the quality of class-conditional samples for non-robust classifiers by providing more stable and informative classifier guidance gradients. The findings highlight the importance of gradient stability in enhancing the performance of classifier guidance, especially on non-robust classifiers.

该研究通过梯度分析比较了鲁棒和非鲁棒分类器，并使用多种梯度稳定技术来提高无条件训练的Denoising Diffusion Probabilistic Model（DDPM）的采样质量。实验结果表明，这些技术通过提供更稳定和信息丰富的分类器指导梯度，显著提高了非鲁棒分类器的类别条件采样质量。研究结果突出了梯度稳定性在增强分类器指导性能方面的重要性，尤其是对于非鲁棒分类器。

GradCheck: 分析分类器引导梯度对条件扩散采样的影响