Stage - Chargé d’étude R&D - amélioration d’une architecture Transformer pour l’exploitation des réseaux électriques F/H
Core
Improve a Transformer architecture to detect critical network elements that could cause fragility in the French high-voltage electricity grid.
Role type
Research intern (Machine Learning)
Builds
Production ML models for grid stability analysis
Domain
Energy / Power Grid / Machine Learning
Deliverable
production ML models
Required skills
Python, PyTorch, Transformer architectures, Hyperparameter optimization, Ablation studies, Classification metrics (PR-AUC)
Preferred skills
Optuna, Electricity grid domain knowledge
Technologies
PyTorch, Optuna
Responsibilities
Understand existing architecture and data; Design evaluation protocols; Conduct ablation studies; Optimize hyperparameters; Evaluate model performance; Document results and prepare open-source code.
Seniority
Intern (Master's/Engineering School level)
