CENTRALE LYON - Postdoctoral Researcher Position Deep Learning for Functional-Oxide Growth Video-to-Spectrum Prediction by RHEED / XRD Fusion
Core
Develop deep learning models to predict X-ray Diffraction (XRD) spectra of functional-oxide thin films from Reflection High-Energy Electron Diffraction (RHEED) videos recorded during Molecular Beam Epitaxy (MBE) growth.
Role type
Postdoctoral Researcher (Deep Learning for Materials Physics)
Builds
A video-to-spectrum prediction model with calibrated uncertainty for in situ control of oxide epitaxy.
Domain
Materials Physics / Artificial Intelligence (Deep Learning)
Deliverable
production ML models
Required skills
Deep learning on sequential/video data, 3D CNNs, temporal transformers, probabilistic models, Python, PyTorch
Preferred skills
Self-supervised video pre-training, uncertainty quantification (Bayesian/ensembles), fuzzy classification, domain adaptation, small-sample/transfer learning
Technologies
PyTorch, TensorFlow, 3D CNN, Vision Transformer, Git
Responsibilities
Build and clean a versioned RHEED/XRD dataset; Pre-train spatio-temporal video encoders; Perform growth regime classification; Design latent representations coupling video encoding with MBE parameters; Learn probabilistic regression of structural parameters and full XRD spectra; Prototype a real-time inference tool.
Seniority
Postdoctoral (2–5 years post-PhD experience)