Denoising Diffusion Autoencoders are Unified Self-supervised Learners

Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. This paper shows that the networks in diffusion models, namely denoising diffusion autoencoders (DDAE), are unified self-supervised learners: by pre-training on unconditional image generation, DDAE has already learned strongly linear-separable representations at its intermediate layers without auxiliary encoders, thus making diffusion pre-training emerge as a general approach for self-supervised generative and discriminative learning. To verify this, we perform linear probe and fine-tuning evaluations on multi-class datasets. Our diffusion-based approach achieves 95.9% and 50.0% linear probe accuracies on CIFAR-10 and Tiny-ImageNet, respectively, and is comparable to masked autoencoders and contrastive learning for the first time. Additionally, transfer learning from ImageNet confirms DDAE's suitability for latent-space Vision Transformers, suggesting the potential for scaling DDAEs as unified foundation models.