Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
This work addresses the challenge of enhancing generative quality in diffusion models for applications like image synthesis, representing an incremental improvement over existing methods.
The paper tackles the problem of improving sample generation in pre-trained diffusion models by introducing Discriminator Guidance, which uses a discriminator to provide explicit supervision on sample realism without joint training, achieving state-of-the-art results on ImageNet 256x256 with FID 1.83 and recall 0.64.
The proposed method, Discriminator Guidance, aims to improve sample generation of pre-trained diffusion models. The approach introduces a discriminator that gives explicit supervision to a denoising sample path whether it is realistic or not. Unlike GANs, our approach does not require joint training of score and discriminator networks. Instead, we train the discriminator after score training, making discriminator training stable and fast to converge. In sample generation, we add an auxiliary term to the pre-trained score to deceive the discriminator. This term corrects the model score to the data score at the optimal discriminator, which implies that the discriminator helps better score estimation in a complementary way. Using our algorithm, we achive state-of-the-art results on ImageNet 256x256 with FID 1.83 and recall 0.64, similar to the validation data's FID (1.68) and recall (0.66). We release the code at https://github.com/alsdudrla10/DG.