Skip to main content

Research & Publications

When other children played with mud, I put it under a microscope.🔬

I discovered my love for science when I was 9 years old. I asked all my relatives and friends to bring me sand samples from vacation and analyzed them with a sieve and microscope, wondering if I could just take any sand and put it into an hourglass. Today, I hold a Bachelor's degree in Geophysics/Oceanography, a Master's degree in Geophysics, as well as a PhD in Machine Learning for Geophysics from the Technical University of Denmark (#103 QS) in collaboration with Heriot-Watt University (#301 QS) in Edinburgh, Scotland.

Iɴ ᴍʏ Bᴀᴄʜᴇʟᴏʀ, I used multi-parameter stacking for seismic interpolation.

Iɴ ᴍʏ Mᴀsᴛᴇʀ, I compared those stacks to diffraction imaging and the standard seismic analysis workflow.

Iɴ ᴍʏ PʜD, I accomplished a few things: I was the first to publish on transfer learning for automatic seismic interpretation. I was part of the team that first tried and published on GANs for seismic inversion. I reviewed the 70-year history of machine learning in geoscience as a book chapter in Advances in Geophysics. I adapted an MIT algorithm for 4D seismic time shift extraction that does not rely on biased ground truth data.

Tᴏᴅᴀʏ I work with deep learning in physical systems with experience in integrating expert knowledge in architectures, heterogeneous data sources, and self-supervised training to avoid label bias. Oh, and I'm a full-time Scientist for Machine Learning at the ECMWF.



2024

  • Matthew Chantry, Mihai Alexe, Simon Lang, Baudouin Raoult, Jesper Dramsch, Florian Pinault, Zied Ben Bouallegue, Mariana Clare, Christian Lessig, Linus Magnusson, and others. Aifs–ecmwf’s data-driven probabilistic forecasting system. In 104th AMS Annual Meeting. AMS, 2024.
    [BibTeX▼]
  • Baudouin Raoult, Matthew Chantry, Florian Pinault, Jesper Dramsh, and Florian Pappenberger. Ai-models: a tool for making forecasts with data-driven nwp models. In 104th AMS Annual Meeting. AMS, 2024.
    [BibTeX▼]

2023

  • Zied Ben-Bouallegue, Mariana C A Clare, Linus Magnusson, Estibaliz Gascon, Michael Maier-Gerber, Martin Janousek, Mark Rodwell, Florian Pinault, Jesper S Dramsch, Simon T K Lang, Baudouin Raoult, Florence Rabier, Matthieu Chevallier, Irina Sandu, Peter Dueben, Matthew Chantry, and Florian Pappenberger. The rise of data-driven weather forecasting. 2023. URL: https://arxiv.org/abs/2307.10128, doi:10.48550/ARXIV.2307.10128.
    [abstract▼] [BibTeX▼]
  • Zied Ben-Bouallegue, Jonathan A Weyn, Mariana C A Clare, Jesper S Dramsch, Peter Dueben, and Matthew Chantry. Improving medium-range ensemble weather forecasts with hierarchical ensemble transformers. 2023. URL: https://arxiv.org/abs/2303.17195, doi:10.48550/ARXIV.2303.17195.
    [abstract▼] [BibTeX▼]

2022

  • F. Vitart, A. W. Robertson, A. Spring, F. Pinault, R. Roškar, W. Cao, S. Bech, A. Bienkowski, N. Caltabiano, E. De Coning, B. Denis, A. Dirkson, J. Dramsch, P. Dueben, J. Gierschendorf, H. S. Kim, K. Nowak, D. Landry, L. Lledó, L. Palma, S. Rasp, and S. Zhou. Outcomes of the WMO prize challenge to improve subseasonal to seasonal predictions using artificial intelligence. Bulletin of the American Meteorological Society, 103(12):E2878–E2886, December 2022. URL: https://doi.org/10.1175/bams-d-22-0046.1, doi:10.1175/bams-d-22-0046.1.
    [abstract▼] [full text] [BibTeX▼]

2021

  • Jesper Sören Dramsch, Mikael Lüthje, and Anders Nymark Christensen. Complex-valued neural networks for machine learning on non-stationary physical data. Computers & Geosciences, 146:104643, 2021.
    [abstract▼] [full text] [BibTeX▼]
  • Runhai Feng, Niels Balling, Dario Grana, Jesper Sören Dramsch, and Thomas Mejer Hansen. Bayesian convolutional neural networks for seismic facies classification. IEEE Transactions on Geoscience and Remote Sensing, pages 1–8, 2021. URL: https://doi.org/10.1109/tgrs.2020.3049012, doi:10.1109/tgrs.2020.3049012.
    [abstract▼] [BibTeX▼]

2020

  • G Corte, Jesper Sören Dramsch, C MacBeth, and H Amini. Deep neural network application for 4d seismic inversion to pressure and saturation: enhancing training data sets. In 82nd EAGE Annual Conference & Exhibition, volume 2020, 1–5. European Association of Geoscientists & Engineers, 2020.
    [abstract▼] [BibTeX▼]
  • Jesper Sören Dramsch. 70 years of machine learning in geoscience in review. In Machine Learning in Geosciences, pages 1–55. Elsevier, 2020. URL: https://doi.org/10.1016/bs.agph.2020.08.002, doi:10.1016/bs.agph.2020.08.002.
    [abstract▼] [BibTeX▼]
  • Gustavo Côrte, Jesper Sören Dramsch, Hamed Amini, and Colin MacBeth. Deep neural network application for 4d seismic inversion to changes in pressure and saturation: optimizing the use of synthetic training datasets. Geophysical Prospecting, 68(7):2164–2185, June 2020. URL: https://doi.org/10.1111/1365-2478.12982, doi:10.1111/1365-2478.12982.
    [abstract▼] [BibTeX▼]
  • Tala Maria Aabø, Jesper Sören Dramsch, Camilla Louise Würtzen, Solomon Seyum, Frédéric Amour, Michael Welch, and Mikael Lüthje. An integrated workflow for fracture characterization in chalk reservoirs, applied to the kraka field. Marine and Petroleum Geology, 2020. URL: http://www.sciencedirect.com/science/article/pii/S026481721930501X, doi:https://doi.org/10.1016/j.marpetgeo.2019.104065.
    [abstract▼] [BibTeX▼]
  • Jesper Sören Dramsch. Machine learning geoscience: applications of deep neural networks in 4d seismic data analysis. PhD Thesis, 2020.
    [abstract▼] [full text] [BibTeX▼]

2019

  • Jesper Sören Dramsch, Anders Nymark Christensen, Colin MacBeth, and Mikael Lüthje. Deep unsupervised 4d seismic 3d time-shift estimation with convolutional neural networks. IEEE Transactions in Geoscience and Remote Sensing, 2019.
    [abstract▼] [full text] [BibTeX▼]
  • Jesper Sören Dramsch, Gustavo Corte, Hamed Amini, Mikael Lüthje, and Colin MacBeth. Deep learning application for 4d pressure saturation inversion compared to bayesian inversion on north sea data. In Second EAGE Workshop Practical Reservoir Monitoring 2019. EAGE Publications BV, 2019. doi:10.3997/2214-4609.201900028.
    [BibTeX▼]
  • Jesper Sören Dramsch, Gustavo Corte, Hamed Amini, Colin MacBeth, and Mikael Lüthje. Including physics in deep learning – an example from 4d seismic pressure saturation inversion. In 81st EAGE Conference and Exhibition 2019 Workshop Programme. EAGE Publications BV, 2019. URL: https://doi.org/10.3997/2214-4609.201901967, doi:10.3997/2214-4609.201901967.
    [abstract▼] [full text] [BibTeX▼]

2018

  • Jesper Sören Dramsch and Mikael Lüthje. Deep-learning seismic facies on state-of-the-art cnn architectures. In SEG Technical Program Expanded Abstracts 2018, 2036–2040. Society of Exploration Geophysicists, 2018. URL: https://doi.org/10.1190/segam2018-2996783.1, doi:10.1190/segam2018-2996783.1.
    [abstract▼] [full text] [BibTeX▼]
  • Lukas Mosser, Wouter Kimman, Jesper Sören Dramsch, Steve Purves, A De la Fuente Briceño, and Graham Ganssle. Rapid seismic domain transfer: seismic velocity inversion and modeling using deep generative neural networks. In 80th EAGE Conference and Exhibition 2018. EAGE Publications BV, 6 2018. URL: https://doi.org/10.3997/2214-4609.201800734, doi:10.3997/2214-4609.201800734.
    [abstract▼] [full text] [BibTeX▼]
  • Jesper Sören Dramsch and Mikael Lüthje. Information theory considerations in patch-based training of deep neural networks on seismic time-series. In First EAGE/PESGB Workshop Machine Learning. EAGE Publications BV, 2018. URL: https://doi.org/10.3997/2214-4609.201803020, doi:10.3997/2214-4609.201803020.
    [abstract▼] [BibTeX▼]
  • Jesper Sören Dramsch, Frédéric Amour, and Mikael Lüthje. Gaussian mixture models for robust unsupervised scanning-electron microscopy image segmentation of north sea chalk. In First EAGE/PESGB Workshop Machine Learning. EAGE Publications BV, 2018. URL: https://doi.org/10.3997/2214-4609.201803014, doi:10.3997/2214-4609.201803014.
    [abstract▼] [BibTeX▼]

2017

  • Tala Maria Aabø, Jesper Sören Dramsch, Michael Welch, and Mikael Lüthje. Correlation of fractures from core, borehole images and seismic data in a chalk reservoir in the danish north sea. In 79th EAGE Conference and Exhibition 2017. EAGE Publications BV, 6 2017. URL: https://doi.org/10.3997/2214-4609.201701283, doi:10.3997/2214-4609.201701283.
    [abstract▼] [full text] [BibTeX▼]

2016

  • Jesper Sören Dramsch. Seismic subsalt imaging with prestack data enhancement methods. Master Thesis, November 2016. URL: https://doi.org/10.31237/osf.io/aec7p, doi:10.31237/osf.io/aec7p.
    [abstract▼] [full text] [BibTeX▼]

2011

  • Jesper Sören Dramsch and DJ Gajewski. Trace interpolation with partial crs stacks. In 73rd EAGE Conference and Exhibition incorporating SPE EUROPEC 2011, cp–238. European Association of Geoscientists & Engineers, 2011.
    [abstract▼] [full text] [BibTeX▼]
  • Jesper Sören Dramsch. Trace interpolation with partial crs-stacks. 2011. URL: https://thesiscommons.org/mvxuh, doi:10.31237/osf.io/mvxuh.
    [abstract▼] [full text] [BibTeX▼]