Description
Training a denoiser on signals gives you a powerful prior over this signal that you can then use to sample examples of this signal.
Papers Using This Method
Whitened Score Diffusion: A Structured Prior for Imaging Inverse Problems2025-05-15Graffe: Graph Representation Learning via Diffusion Probabilistic Models2025-05-08Score-based Self-supervised MRI Denoising2025-05-08Generalization through variance: how noise shapes inductive biases in diffusion models2025-04-16Generalization error bound for denoising score matching under relaxed manifold assumption2025-02-19Concentration Inequalities for the Stochastic Optimization of Unbounded Objectives with Application to Denoising Score Matching2025-02-12Can Diffusion Models Learn Hidden Inter-Feature Rules Behind Images?2025-02-07Sequential Change Point Detection via Denoising Score Matching2025-01-22Semi-Implicit Functional Gradient Flow for Efficient Sampling2024-10-23Classification-Denoising Networks2024-10-04What's the score? Automated Denoising Score Matching for Nonlinear Diffusions2024-07-10Evaluating the design space of diffusion-based generative models2024-06-18DualBind: A Dual-Loss Framework for Protein-Ligand Binding Affinity Prediction2024-06-11Nonlinear denoising score matching for enhanced learning of structured distributions2024-05-24SocialGFs: Learning Social Gradient Fields for Multi-Agent Reinforcement Learning2024-05-03Structure-Guided Adversarial Training of Diffusion Models2024-02-27Label-Noise Robust Diffusion Models2024-02-27Target Score Matching2024-02-13Time Series Diffusion in the Frequency Domain2024-02-08Analyzing Neural Network-Based Generative Diffusion Models through Convex Optimization2024-02-03