Complex-valued neural networks for machine learning on non-stationary physical data

Autor: Dramsch, Jesper Sören, Lüthje, Mikael, Christensen, Anders Nymark

Deep learning has become an area of interest in most scientific areas, including physical sciences. Modern networks apply real-valued transformations on the data. Particularly, convolutions in convolutional neural networks discard phase information e

Externí odkaz: http://arxiv.org/abs/1905.12321

Complex-valued neural networks for machine learning on non-stationary physical data

Autor: Dramsch, Jesper Sören, Lüthje, Mikael, Christensen, Anders Nymark

Publikováno v: In Computers and Geosciences January 2021 146

Low-Cost CCS Monitoring with Sparse Data Acquisition

Autor: Mohammadzaheri, Afsaneh, de Ridder, Sjoerd, Calvert, Alex, Lüthje, Mikael

Publikováno v: Mohammadzaheri, A, de Ridder, S, Calvert, A & Lüthje, M 2022, ' Low-Cost CCS Monitoring with Sparse Data Acquisition ', Danish Offshore Technology Conference 2022, Kolding, Denmark, 29/11/2022-30/11/2022 .

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______1202::7c12c33dba2793c4898eef1fce88ecc4
https://orbit.dtu.dk/en/publications/1e5bfdd8-10db-4f28-a29a-d35f2928a8ae

DFN Generator v2.0: A new tool to model the growth of large-scale natural fracture networks using fundamental geomechanics.

Autor: Welch, Michael J., Lüthje, Mikael, Oldfield, Simon J.

Publikováno v: Geoscientific Model Development Discussions; 5/19/2022, p1-42, 42p

Information Theory Considerations In Patch-Based Training Of Deep Neural Networks On Seismic Time-Series

Autor: Dramsch, Jesper Soeren, Lüthje, Mikael

Recent advances in machine learning relies on convolutional deep neural networks. These are often trained on cropped image patches. Pertaining to non-stationary seismic signals this may introduce low frequency noise and non-generalizability.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c6fa55bada144b7f72bde70822ddb8b