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...
Hlavní autoři: | De Bortoli, V, Mathieu, E, Hutchinson, M, Thornton, J, Teh, YW, Doucet, A |
---|---|
Médium: | Conference item |
Jazyk: | English |
Vydáno: |
Curran Associates
2023
|
Podobné jednotky
-
Diffusion Schrödinger bridge with applications to score-based generative modeling
Autor: De Bortoli, V, a další
Vydáno: (2022) -
Generative models as distributions of functions
Autor: Dupont, E, a další
Vydáno: (2022) -
Riemannian geodesics of semi Riemannian warped product metrics
Autor: Oriella M. Amici, a další
Vydáno: (2013-01-01) -
Riemannian geometry /
Autor: Gallot, S. (Sylvestre), 1948-, a další
Vydáno: (2004) -
Dynamically generated inflation from non-Riemannian volume forms
Autor: D. Benisty, a další
Vydáno: (2019-09-01)