PhD Position F/M Pretrained models of multimodal neuroimaging for predicting individual cognition

Develop pretrained models of functional, diffusion and structural MRI that predict individual cognition with calibrated uncertainty. The ANR-funded MIND–GIN Bordeaux collaboration combines preprocessing and harmonization of large datasets, generative and self-supervised models, and transfer to smaller cognitive-neuroscience studies. The doctorate will release open pipelines and benchmarks. Python and PyTorch or JAX, probability, statistics and representation learning are required; neuroimaging and large-scale computing experience are desirable.

Appointment: 3 years; advertised start 2027-01-01. 2300€ gross/month.

Graduate degree or equivalent. Applicants may require authorization for access to a restricted research area (ZRR).