SDSC Hub at PSI

The Swiss Data Science Center (SDSC) is a strategic initiative of the ETH Domain, with EPFL and ETH Zurich as founding partners. Its mission is to accelerate the adoption of data science, machine learning and artificial intelligence (AI) within the Swiss academic community, the industrial sector, and public institutions, to enable data-driven science and innovation for societal impact. With a multidisciplinary team of data scientists, computing specialists and domain experts located in Zürich (ETH), Lausanne (EPFL), and Villigen (PSI), the SDSC engages in collaborative research and provides expertise and services addressing a wide range of needs in various sectors, such as in biomedical data science, energy, climate and environment, digital administration, engineering and architecture, large-scale computing infrastructures or social sciences. More information: www.datascience.ch 

The SDSC office at PSI is primarily responsible for developing advanced data science (ML/AI) solutions to address scientific questions as well as associated operational needs related to the large-scale research infrastructures that are strategic for the competitive advantage of Switzerland in the research and innovation landscape.

With a team of experienced data scientists with broad expertise in ML and AI, our research primarily focuses on the development of novel representations and sampling strategies for efficient data intake and smart experimentation, generative AI for inverse problems in (biomedical) imaging, and time-series analysis.

Research Interests
  • Machine Learning
  • Imaging and Inverse Problems
  • Signal Processing and Optimization
  • Time-series and Experimental Design
  • Gasparotto P, Barba L, Stadler HC, Assmann G, Mendonça H, Ashton AW, et al.
    TORO Indexer: a PyTorch-based indexing algorithm for kilohertz serial crystallography
    Journal of Applied Crystallography. 2024; 57(4): 931-944. https://doi.org/10.1107/S1600576724003182
    DORA PSI
  • Koka T, Tsakiris MC, Muma M, Béjar Haro B
    Shuffled multi-channel sparse signal recovery
    Signal Processing. 2024; 224: 109579 (14 pp.). https://doi.org/10.1016/j.sigpro.2024.109579
    DORA PSI
  • Tang Y, Béjar B, Vidal R
    Semantic-aware video representation for few-shot action recognition
    In: 2024 IEEE winter conference on applications of computer vision (WACV 2024). Proceedings. Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE); 2024:6444-6454. https://doi.org/10.1109/WACV57701.2024.00633
    DORA PSI
  • Ansuinelli P, Béjar Haro B, Ekinci Y, Mochi I
    Towards fast ptychography image reconstruction of EUV masks by deep neural networks
    In: Liang T, ed. Photomask technology 2023. Vol. 12751. Proceedings of SPIE. SPIE; 2023:127510Q (10 pp.). https://doi.org/10.1117/12.2685227
    DORA PSI
  • Sherratt K, Gruson H, Grah R, Johnson H, Niehus R, Prasse B, et al.
    Predictive performance of multi-model ensemble forecasts of COVID-19 across European nations
    eLife. 2023; 12: e81916 (19 pp.). https://doi.org/10.7554/eLife.81916
    DORA PSI
  • Tang Y, Bejar B, Essoe JK-Y, McGuire JF, Vidal R
    Facial tic detection in untrimmed cideos of Tourette Syndrome patients
    In: 2022 26th international conference on pattern recognition (ICPR). International conference on pattern recognition. sine loco: IEEE; 2022:3152-3159. https://doi.org/10.1109/ICPR56361.2022.9956140
    DORA PSI