Predicting Psychopathology Factors for the NeurIPS 2025 EEG Foundation Challenge

! POSITION FILLED !

Download the internship proposal

Stage type

Master 2 Computer science

Date and duration of the internship

5 months, second semester 2025, approximately from aug to dec (plus-minus one month)

Context

This 5-month internship project is centered on « Challenge 2: Decoding psychopathology factor from the EEG » of the prestigious NeurIPS 2025 EEG Foundation Challenge. The primary objective is to develop a novel deep learning model capable of regressing continuous psychopathology scores (e.g., p-factor, internalization, attention) from high-density EEG data.

The intern will be the lead coder for this project, tasked with implementing, training, and optimizing models using the large-scale Healthy Brain Network (HBN-EEG) dataset, which contains recordings from over 3,000 individuals. They will operate within a small, agile team of PhD and Master’s students who will provide intermittent support and collaborative input. The project culminates in a code submission to the official competition, aiming for a high rank on the leaderboard. The student will work using Jean Zay resources already obtained for this project.

This initiative directly addresses a critical goal in clinical neuroscience: the development of objective, EEG-based biomarkers for mental health to augment traditional subjective assessments.

Objective

Develop a flexible deep learning pipeline to solve the challenges in the competition, specifically Challenge 2.

Missions

the intern will
● Read the literature on EEG transfer learning
● Implement the pipeline
● gain extensive, hands-on experience in applying advanced machine learning and deep learning techniques to complex, high-dimensional time-series data
● Boost their career by participating in a major machine learning conference competition
● Experience in BCI data processing
● Experience working in team
● Experience working on a big codebase

Skills required

strong programming skills, Python, deep learning (pytorch), good level of English

Contact

Dmitrii Todorov, PhD; Chaire Professeur Junior
dmitrii.todorov[at]inserm.fr

Host laboratory

Laboratoire d’Imagerie Biomédicale, LIB, Sorbonne Université, Campus des Cordeliers, 15 rue de l’École de Médecine, 75006 Paris.