Postdoc in Uncertainty-Aware Machine Learning Force Fields

About us

The University of Luxembourg is an international research university with a distinctly multilingual and interdisciplinary character.

The Faculty of Science, Technology and Medicine (FSTM) at the University of Luxembourg contributes multidisciplinary expertise in the fields of Mathematics, Physics, Engineering, Computer Science, Life Sciences and Medicine.
Through its dual mission of teaching and research, the FSTM seeks to generate and disseminate knowledge and train new generations of responsible citizens in order to better understand, explain and advance society and environment we live in.

Your role

We invite applications for a postdoctoral researcher to join the UMLFF project at the University of Luxembourg. The project aims to develop the next generation of uncertainty-aware machine-learning force fields (MLFFs) that combine state-of-the-art equivariant neural network architectures with robust, well-calibrated uncertainty estimates. These models will enable fully automated active learning in configurational and chemical space, paving the way for general-purpose MLFFs that are both accurate and reliable across diverse chemical systems.

The postdoctoral researcher will work closely with Dr. Igor Poltavskyi, a leading expert in MLFFs, and within the Theoretical Chemical Physics group led by Prof. Alexandre Tkatchenko, one of the world's leading groups in machine-learning force-field development. The project benefits from access to state-of-the-art HPC resources, including the MeluXina supercomputer, and from strong international collaborations, e.g., Max Planck Institute, Google DeepMind, TU Berlin.

The successful candidate will contribute to the development of uncertainty-aware foundational MLFFs based on modern SO(3)-equivariant message-passing neural networks, such as MACE and SO3LR. Core scientific themes include:

• Intrinsic uncertainty estimation in MLFFs (epistemic & aleatoric uncertainty)
• Negative log-likelihood and calibration methods for force-field training
• Feature-space and orbit-based analysis of atomic environments
• Detection of extrapolation and low-reference data regimes
• Active learning in configurational and chemical space
• Training and benchmarking of large-scale foundational MLFF models

More precisely, the postdoctoral researcher will:

• Develop and implement uncertainty-aware MLFF architectures
• Design and test loss functions, calibration methods, and training strategies
• Analyze feature-space structure, atomic orbits, and model applicability domains
• Train and benchmark large-scale MLFF models on diverse molecular and materials datasets
• Integrate uncertainty estimates into active learning pipelines
• Publish results in leading journals and present them at international conferences
• Collaborate with PhD students and international research partners

Your profile

Required qualifications:

• PhD in physics, chemistry, materials science, computational science, or a related field
• Strong background in machine learning and/or computational chemistry
• Experience with deep neural networks and scientific programming with Python
• Solid understanding of molecular simulations, force fields, or atomistic modeling

Highly desirable:

• Experience with equivariant or graph neural networks, e.g., NequIP, MACE, SO3LR, Allegro
• Experience with MLFF training and validation
• Knowledge of uncertainty estimation, Bayesian or ensemble methods
• Experience with HPC environments and large-scale model training
• Familiarity with DFT codes, e.g., FHI-aims, and molecular dynamics

We offer

• Multilingual and international character. Modern institution with a personal atmosphere. Staff coming from 90 countries. Member of the "University of the Greater Region" (UniGR)
• A modern and dynamic university. High-quality equipment. Close ties to the business world and to the Luxembourg labour market. A unique urban site with excellent infrastructure
• A partner for society and industry. Cooperation with European institutions, innovative companies, the Financial Centre and with numerous non-academic partners such as ministries, local governments, associations, NGOs …

How to apply

Applications should include:

Early application is highly encouraged, as the applications will be processed upon reception. Please apply ONLINE formally through the HR system. Applications by Email will not be considered.

All qualified individuals are encouraged to apply. In line with our values, the University of Luxembourg promotes an inclusive culture. We encourage applications from individuals of all backgrounds and are dedicated to upholding equality and respect for our employees and students.

General information:

• Contract Type: Fixed Term Contract 12 Month
• Work Hours: Full Time 40.0 Hours per Week
• Location: Limpertsberg Campus
• Internal Title: Postdoctoral researcher
• Job Reference: UOL07996

The yearly gross salary for every Postdoctoral Researcher at the UL is EUR 85176 (full time).