EEG Foundation Challenge Start Kits

This repository contains start kits for the EEG Foundation challenges, a NeurIPS 2025 competition focused on advancing EEG decoding through cross-task transfer learning and psychopathology prediction.

🚀 Quick Start

Challenge 1: Cross-Task Transfer Learning

Goal: Develop models that can effectively transfer knowledge from passive EEG tasks to active cognitive tasks.

Challenge 2: Predicting the P-factor from EEG

Goal: Predict the psychopathology factor (P-factor) from EEG recordings to enable objective mental health assessments.

📁 Repository Structure

Main Files

• challenge_1.ipynb - Complete tutorial for Challenge 1: Cross-task transfer learning

  • Understanding the Contrast Change Detection (CCD) task
  • Loading and preprocessing EEG data using EEGDash
  • Building deep learning models with Braindecode
  • Training and evaluation pipeline

• challenge_1.py - Python script version of Challenge 1 notebook for easier integration

• challenge_2.ipynb - Tutorial for Challenge 2: P-factor regression

  • Understanding the P-factor regression task
  • Data loading and windowing strategies
  • Model training for psychopathology prediction

• challenge_2.py - Python script version of Challenge 2 notebook for easier integration

• submission.py - Template for competition submission

  • Shows required format for model submission
  • Includes examples for both challenges

• requirements.txt - Python dependencies needed to run the notebooks

Advanced Examples (not_ready_yet/)

• challenge_2_self_supervised.ipynb - Advanced self-supervised learning approach
  • Implementing Relative Positioning (RP) for unsupervised representation learning
  • Fine-tuning for P-factor prediction
  • PyTorch Lightning integration
  • Note: This is an advanced example that may require additional setup

🛠️ Installation

pip install -r requirements.txt

Main dependencies:

• braindecode - Deep learning library for EEG
• eegdash - Dataset management and preprocessing
• pytorch - Deep learning framework

🤝 Community & Support

This is a community competition with a strong open-source foundation. If you see something that doesn't work or could be improved:

1. Please be kind - we're all working together
2. Open an issue in the issues tab
3. Join our weekly support sessions (starting 08/09/2025)

The entire decoding community will only go further when we stop solving the same problems over and over again, and start working together!

📚 Resources

• Competition Website
• EEGDash Documentation
• Braindecode Models
• Dataset Download Guide