Stage - Explainable AI (xAI) for Industrial Applications: Literature Review and Implementation on Real Use Cases - H/F
Core
Mapping and implementing explainable AI (xAI) methods to interpret industrial machine learning models for time series, computer vision, and hybrid physical-ML systems.
Role type
Research intern (Master 2 or Engineering School)
Builds
Reproducible code and evaluation frameworks for xAI methods across three industrial use cases.
Domain
Industrial AI, Machine Learning, Explainable AI
Deliverable
production ML models
Required skills
Machine learning, Statistics, Python programming, Time series analysis, Neural network architectures (ANN, CNN, transformers, ViT), Explainability techniques (SHAP, LIME, Grad-CAM, Integrated Gradients)
Preferred skills
Scientific article synthesis, Technical communication
Technologies
Python, scikit-learn, PyTorch
Responsibilities
Conduct literature review on xAI methods, Map and catalog xAI tools and evaluation metrics, Apply and compare xAI methods on real industrial datasets, Validate explanations with domain experts, Present findings via reports and demos
