5 papers
Diffusion models recover accurate mixture weights despite score function insensitivity
Andrew Dennehy, Ramchandran Muthukumar, Rebecca Willett +1
Score-based generative models exhibit a puzzling behavior: they often appear to cover all modes of a target multimodal distribution and yet may fail to learn the correct relative m…
From Spectral Methods to Sample Complexity Bounds for Fourier Neural Operators
Nisha Chandramoorthy, Daniel Sanz-Alonso, Nathan Waniorek
We establish approximation and learning guarantees for Fourier neural operators (FNOs) applied to time- solution operators of dissipative evolution equations. The analysis build…
On the Robustness of Distribution Support under Diffusion Guidance
Ruijia Cao, Yuchen Wu, Nisha Chandramoorthy
Diffusion guidance is a powerful technique that enables controllable and high-fidelity sample generation with diffusion models. At a high level, it modifies the score function by i…
When and how can inexact generative models still sample from the data manifold?
Nisha Chandramoorthy, Adriaan de Clercq
A curious phenomenon observed in some dynamical generative models is the following: despite learning errors in the score function or the drift vector field, the generated samples a…
When are dynamical systems learned from time series data statistically accurate?
Jeongjin Park, Nicole Yang, Nisha Chandramoorthy
Conventional notions of generalization often fail to describe the ability of learned models to capture meaningful information from dynamical data. A neural network that learns comp…