4 papers
Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
Naïl B. Khelifa, Richard E. Turner, Ramji Venkataramanan
Score-based diffusion models are typically trained by minimizing the score matching error, and standard theoretical analyses rely on this quantity to bound the sampling discr…
Recursively Trained Diffusion Models: Limiting Collapse Distribution and Spectral Characterization
Naïl B. Khelifa, Richard E. Turner, Ramji Venkataramanan
Recursive training of generative models on their own outputs can lead to model collapse, a compounding drift away from the true data distribution. Existing theoretical works bound…
Quantifying Error Propagation and Model Collapse in Diffusion Models
Nail B. Khelifa, Richard E. Turner, Ramji Venkataramanan
Machine learning models are increasingly trained or fine-tuned on synthetic data. Recursively training on such data has been observed to significantly degrade performance in a wide…
Enhanced Denoising and Convergent Regularisation Using Tweedie Scaling
Naïl Khelifa, Ferdia Sherry, Carola-Bibiane Schönlieb
The inherent ill-posed nature of image reconstruction problems, due to limitations in the physical acquisition process, is typically addressed by introducing a regularisation term…