Riemannian score-based generative modelling
Score-based generative models (SGMs) are a powerful class of generative models that exhibit remarkable empirical performance.Score-based generative modelling (SGM) consists of a noising'' stage, whereby a diffusion is used to gradually add Gaussian noise to data, and a generative model, wh...
Главные авторы: | De Bortoli, V, Mathieu, E, Hutchinson, M, Thornton, J, Teh, YW, Doucet, A |
---|---|
Формат: | Conference item |
Язык: | English |
Опубликовано: |
Curran Associates
2023
|
Схожие документы
-
Diffusion Schrödinger bridge with applications to score-based generative modeling
по: De Bortoli, V, и др.
Опубликовано: (2022) -
Generative models as distributions of functions
по: Dupont, E, и др.
Опубликовано: (2022) -
Riemannian geodesics of semi Riemannian warped product metrics
по: Oriella M. Amici, и др.
Опубликовано: (2013-01-01) -
Riemannian geometry /
по: Gallot, S. (Sylvestre), 1948-, и др.
Опубликовано: (2004) -
Dynamically generated inflation from non-Riemannian volume forms
по: D. Benisty, и др.
Опубликовано: (2019-09-01)