postdoctoral position in deep learning for photometric redshifts

 Applications are invited for a postdoctoral position at the Laboratoire d’Astrophysique de Marseille (LAM), France, to work on deep learning methods for photometric redshift estimates with large imaging surveys. The position is funded by the French National Research Agency (“DEEPDIP” project). The appointment is for 2 years, starting mid 2020.
  The project is a collaboration between CCPM (Marseille), TETIS (Montpellier) and IAP (Paris). The candidate is expected to develop deep learning techniques, in particular to handle the unbalanced and incomplete representativity of the spectroscopic training sets currently plaguing machine learning methods. S/he will have the opportunity to participate in and initiate scientific analyses using the resulting millions of photometric redshifts estimated on current large imaging surveys (CFHTLS, KIDS, HSC, …). Such work will be a stepping stone to the LSST survey.
Required qualifications:  Ph.D. or equivalent degree in physics, astronomy or data science; experience in Machine Learning/Deep Learning.
LAM has an animated research atmosphere in various fields of astrophysics. Marseille is a mediterranean harbor with a colorful ambience and cultural mix.
Applicants should send their CV, research summary and a letter of interest for the position to S. Arnouts ([email protected]), as well as three reference letters.