Bandeau image

Call for Thesis Projects [UDOPIA Programme]

Call for Thesis Projects [UDOPIA Programme]

Chapo
Launched in 2020, the UDOPIA PhD program in AI draws on the considerable strengths of the University of Paris-Saclay and related fields to create a unique cohort of PhD students trained at the forefront of key AI topics.
Contenu
Corps de texte

Students will benefit from a rich ecosystem with close links to industry, and from existing instruments such as the DATAIA Paris-Saclay Institute or the SaclayIA computing platform. Specific accompanying measures will promote interdisciplinarity, mobility, entrepreneurship and the wide dissemination of research results to universities, industry and the general public.

The doctoral programme is co-financed by the French National Research Agency (ANR), and by the University of Paris-Saclay, its components, associates and partners, notably CentraleSupelec, CMLA's IDAML Chair, DATAIA, ENS Paris-Saclay, INRIA Saclay, Labex Hadamard, UEVE, Vedecom.

The programme consists of three calls for thesis projects in 2020, 2021 and 2022. 

Call 2022

Discover the text of the 2022 call

PhD students selected in 2022 will be recruited on a 36-month fixed-term contract starting in October 2022, with a gross monthly salary of €1925.

Results of the UDOPIA call

List of winners

  • Zhe ZHENG / Doctoral School EDMH - Modeling uncertainty in recurrent neural networks for efficient video restoration
  • Julia LASCAR / Doctoral School STIC - Machine learning for multi/hyperspectral data fusion, application to X-ray imaging in astrophysics
  • Clément BERNARD / Doctoral School STIC - Computational methods based on deep learning for the prediction of 3D structures of RNA
  • Maja PANTIĆ / Doctoral School EOBE - Hybrid and deep learning architectures for quantitative susceptibility imaging in the human brain at 7 Tesla
  • Adrien CANCÈS / Doctoral School EDMH - Particle-based methods for high-dimensional multi-marginal optimal transport
  • Ambroise HEURTEBISE / Doctoral School STIC - Multi-view learning: from ICA to self-supervision
  • Zixing QIU / Doctoral School 2MIB - Deep learning for molecule identification in the interstellar medium