no code implementations • 9 Jul 2023 • Giovanni Angelo Meles, Macarena Amaya, Shiran Levy, Stefano Marelli, Niklas Linde
By sampling a low-dimensional prior probability distribution defined in the low-dimensional latent space of such a model, it becomes possible to efficiently sample the physical domain at the price of a generator that is typically highly non-linear.
no code implementations • 8 Sep 2021 • Cédric Travelletti, David Ginsbourger, Niklas Linde
Furthermore, in that context, covariance matrices can become too large to be stored.
no code implementations • 16 Apr 2021 • Eliane Maalouf, David Ginsbourger, Niklas Linde
We propose a novel approach for solving inverse-problems with high-dimensional inputs and an expensive forward mapping.
1 code implementation • 21 Dec 2018 • Eric Laloy, Niklas Linde, Cyprien Ruffino, Romain Hérault, Gilles Gasso, Diedrik Jacques
Global probabilistic inversion within the latent space learned by Generative Adversarial Networks (GAN) has been recently demonstrated (Laloy et al., 2018).
Geophysics
no code implementations • 25 Oct 2017 • Eric Laloy, Romain Hérault, John Lee, Diederik Jacques, Niklas Linde
Here, we use a deep neural network of the variational autoencoder type to construct a parametric low-dimensional base model parameterization of complex binary geological media.
no code implementations • 16 Aug 2017 • Eric Laloy, Romain Hérault, Diederik Jacques, Niklas Linde
After training, realizations containing a few millions of pixels/voxels can be produced in a matter of seconds.