PhD in computational neuroscience

KEY WORDS

CITY

COUNTRY

France

DETAILS OF THE OFFER

Working place: CERCO (Brain and Cognition Research Center)

Missions:

The project builds upon a pilot study, in which participants listen to speech in a MEG scanner, and Ridge regressions were trained to reconstruct the spectrogram of the speech from the MEG signal (similar to [3]). This study showed that a poorly reconstructed frequency band often indicates that this band is not well perceived. Yet Ridge regressions could easily miss relevant neural patterns that are useful to reconstruct certain bands accurately. If this happened, we would incorrectly conclude a hearing deficit in this band. The main goal of the internship it to replace the Ridge regressions with deep neural networks (DNN), which are much more expressive and thus less likely to miss such informative patterns. Thus, an inaccurate DNN-based reconstruction in a certain frequency band will more reliably indicate a hearing deficit.

Qualifications:

The candidate should be willing to work in a multidisciplinary environment which combines AI and neurosciences, have a good level in English, excellent mathematical and programming skills and a keen interest in the brain.

+infos

TYPE OF JOB

TYPE OF CONTRACT

REMUNERATION

APPLICATION DEADLINE

EMPLOYMENT START DATE