DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Models
This addresses the challenge of explainable factor discovery in generative models for researchers and practitioners in machine learning, representing a novel method for a known bottleneck.
The paper tackles the problem of automatically discovering and disentangling the underlying factors in Diffusion Probabilistic Models (DPMs) without annotations, proposing DisDiff to achieve this and demonstrating its effectiveness on synthetic and real-world datasets.
Targeting to understand the underlying explainable factors behind observations and modeling the conditional generation process on these factors, we connect disentangled representation learning to Diffusion Probabilistic Models (DPMs) to take advantage of the remarkable modeling ability of DPMs. We propose a new task, disentanglement of (DPMs): given a pre-trained DPM, without any annotations of the factors, the task is to automatically discover the inherent factors behind the observations and disentangle the gradient fields of DPM into sub-gradient fields, each conditioned on the representation of each discovered factor. With disentangled DPMs, those inherent factors can be automatically discovered, explicitly represented, and clearly injected into the diffusion process via the sub-gradient fields. To tackle this task, we devise an unsupervised approach named DisDiff, achieving disentangled representation learning in the framework of DPMs. Extensive experiments on synthetic and real-world datasets demonstrate the effectiveness of DisDiff.