Postdoctoral Position in Dynamical Image Restoration for Radio Interferometry at LESIA

Annonce transmise par Bapiste Cecconi (LESIA)


Join the EXTRACT project (EU, in designing an edge-to-cloud solution for heavy data processing, employing Deep Learning methodologies. The project includes a pilot named Transient Astrophysics using SKA pathfinders (TASKA), focusing on dynamical astronomical imaging data processing in radio using DL techniques.
Specific Problematic:
Radio astronomy imaging poses challenges in imaging dynamical events in the sky due to traditional methods averaging out short-lived events or poorly imaging extended time-variable emissions. Leveraging machine learning and deep learning, we aim to revisit imaging and deconvolution techniques to produce unbiased image cubes while preserving physical information. Image restoration techniques developed in other domains will be applied to the project, focusing on imaging the Sun as a variable and extended radio source.
Role Overview:
As a Postdoctoral Researcher, you will contribute to the development of deep learning networks for imaging and deconvolution of radio interferometry data. Your work will involve modeling varying sources as 4D structured signals to detect and restore astrophysical transients. You will utilize both simulated and real data for training sets.
Detailed info:
Application Process:
To apply, please submit your CV, cover letter, and contact information for three references to Baptiste Cecconi ( and Julien Girard ( Review of applications will begin May 15th 2024 and continue until the position is filled.
Additional Information:
- Position Type: Full-time, Postdoctoral Researcher
- Location: LESIA, Observatoire de Paris, Meudon , France
- Duration: 18 months
- Start Date: as soon as possible
For inquiries about the position, please contact Baptiste Cecconi ( and Julien Girard (