Doktorand i multimodal maskininlärning för energilagringsmaterial
Core
Develop a multimodal machine learning framework integrating heterogeneous data streams from experiments and simulations to model electrochemical interfaces in batteries.
Role type
PhD researcher (Probabilistic Machine Learning & Data Assimilation)
Builds
Interpretable, uncertainty-quantified models of interface formation and growth in Li-, Zn-, and Cu-based battery systems.
Domain
Energy storage materials / Electrochemistry / Scientific Computing
Deliverable
production ML models
Required skills
Probabilistic machine learning, Bayesian inference, Data assimilation, Deep state-space models, Scientific computing, Python programming
Preferred skills
Generative models, PyTorch or JAX, Electrochemistry, X-ray scattering techniques, Time-series data analysis
Technologies
Python, PyTorch, JAX, Lattice-Boltzmann, Phase-field simulations
Responsibilities
Build and validate inference and simulation pipelines, Participate in planning and execution of operando experiments (including synchrotron access at PETRA III), Develop expertise in probabilistic ML and data assimilation, Analyze learned latent spaces to extract physical meaning.
Seniority
PhD Candidate (Research focus, max 20% teaching)