Organisation/Company: CRAN (University of Lorraine and CNRS)
Research Field: Engineering » Electrical engineering
Researcher Profile: Recognised Researcher (R2)
Country: France
Application Deadline: 31 Dec 2024 - 00:00 (Africa/Abidjan)
Type of Contract: Temporary
Job Status: Full-time
Is the job funded through the EU Research Framework Programme? Not funded by a EU programme
Is the Job related to staff position within a Research Infrastructure? No
Offer Description We are offering a postdoc position on the development of statistical and tensor decomposition methods for representation learning of heterogeneous data with application to the analysis of neuroimaging data.
Location: The CRAN laboratory (University of Lorraine) at Nancy, France, with visits to the MLSP laboratory (UMBC) in Maryland, USA. The candidate will work with Prof. Sebastian Miron, Dr. Ricardo Borsoi, and Prof. David Brie in the CRAN laboratory, Nancy, and with Prof. Tülay Adali at the MLSP laboratory, UMBC, USA.
The starting date is flexible (the position is open until filled).
Description: The analysis of spatiotemporal data is a fundamental problem in multiple domains such as neuroscience, epidemiology, climate science, and pollution monitoring. Developing representation learning methods for spatiotemporal data that can effectively and jointly handle data from diverse modalities poses a significant challenge. A particular difficulty is to devise flexible models which are directly interpretable, readily providing insight into the relationships that are learned from the data. The candidate will develop flexible representation learning and data analysis methods specifically designed to handle heterogeneous spatiotemporal data, effectively utilizing both algebraic (matrix and tensor decompositions) and statistical frameworks to generate results that are interpretable and backed by statistical guarantees. The developed methods will be applied to personalized medicine, with the aim to elucidate the interplay between neuroimaging data (e.g., fMRI) and cognitive/socioeconomic factors as well as their temporal evolution.
Candidate profile: Ph.D. degree in signal processing, machine learning, applied mathematics, or related fields.
#J-18808-Ljbffr