Annonce en lien avec l’Action/le Réseau : aucun
Laboratoire/Entreprise : LIS UMR 7020
Durée : 2 years
Contact : francois-xavier.dupe@lis-lab.fr
Date limite de publication : 2018-05-11
Contexte :
In brain imaging, traditional group analyses rely on averaging data collected in different individuals. This averaging offers a summary representation of the studied group, thus providing a way to perform inference at the population level. However, it discards the specificities of each individual, which have recently proved to carry critical information to develop diagnosis and prognosis tools for neurological and psychiatric diseases or to understand high level cognitive processes.
Sujet :
Estimating robust population-wise invariants while preserving individual specificities is a challenge that can be addressed by integrating the information offered by different neuroimaging modalities, such as anatomical, functional and diffusion MRI, which respectively allow assessing brain shape, activity and connectivity. This can therefore be framed as a multi-view machine learning question. The tasks of the post-doctoral fellow will consist in 1. finding adequate representations of data (e.g. graph, stack of images, …) that preserve structural information, 2. designing and implementing machine learning algorithms that exploit both the representations and the multiple views using kernel methods and/or neural networks, and 3. evaluating them on a variety of MRI datasets dedicated to studying language and communication.
Profil du candidat :
The candidate should have completed a PhD in computer science, applied mathematics or electrical engineering, with a focus on machine learning. He/she should also have a strong motivation to work in neuroscience, as the working environment will be truly inter-disciplinary. The two years postdoctoral fellowship, funded by the newly established Institute for Language, Communication and the Brain (http://www.ilcb.fr) will be awarded through a competitive selection process. Interested candidates should contact sylvain.takerkart@univ-amu.fr, francois-xavier.dupe@lis-lab.fr and hachem.kadri@lis-lab.fr before May 11 2018.
Formation et compétences requises :
The candidate will be fluent in Python, have a good knowledge about recent methods in Machine Learning (e.g. kernels methods, neural network…). An interest in methods dealing with graphs like graph kernels will be appreciate.
Adresse d’emploi :
The post-doc will take place in the LIS offices in Marseille.