Job Information
- Organisation/Company: University of Latvia
- Department: Mathematics
- Research Field: Mathematics » Applied mathematics; Mathematics » Computational mathematics; Physics » Mathematical physics; Physics » Computational physics
Offer Description
The University of Latvia is offering a funded PhD position in the area of physics-informed neural operators for modelling complex physical processes governed by partial differential equations.
The PhD project combines applied mathematics, deep learning, and scientific computing, with a strong emphasis on operator learning methods such as DeepONet and Fourier Neural Operators, including their physics-informed extensions. The research will focus on developing theoretical understanding and practical improvements of neural operators, with direct relevance to challenging applications in hydrodynamics, in particular crystal growth processes and wave dynamics in coastal and harbour environments.
The work is theory-driven but closely connected to applications: the goal is to develop mathematical and algorithmic insights that improve robustness, generalization, and physical consistency of neural operators used as fast surrogate models for PDE-based simulations.
We are particularly interested in candidates with a background in applied mathematics, especially those working at the interface of mathematical modelling or theoretical physics, with a solid understanding of PDEs and numerical methods, and some prior experience in deep learning (e.g. training neural networks, familiarity with modern ML frameworks).
The PhD position is funded for up to three years. The project-funded salary is approximately 2000 EUR per month before tax and corresponds to the funded workload within this project rather than a full-time appointment. Depending on the candidate’s profile and interests, there may also be opportunities for additional paid involvement in related machine learning and AI-for-science projects at the University of Latvia, as well as participation in international research collaborations.
Where to apply
E-mail: janis.bajars@lu.lv
Requirements
Research Field: Mathematics » Applied mathematics — Education Level: Master Degree or equivalent
Research Field: Mathematics » Computational mathematics — Education Level: Master Degree or equivalent
Research Field: Physics » Computational physics — Education Level: Master Degree or equivalent
Research Field: Physics » Mathematical physics — Education Level: Master Degree or equivalent
Skills/Qualifications
A Master’s degree in applied mathematics, mathematical physics, computational science, or a related field; strong knowledge of partial differential equations and numerical methods; experience in mathematical modelling and scientific computing; prior experience with machine learning/deep learning and modern ML frameworks (e.g., PyTorch, TensorFlow, or JAX); strong programming and analytical skills. Knowledge or experience in physics-informed machine learning, neural operators (e.g., DeepONet, Fourier Neural Operators), hydrodynamics, wave dynamics, or related areas is an advantage. Excellent English communication skills and the ability to conduct independent research are expected.
Languages: ENGLISH — Level: Excellent
Additional Information
Work Location(s)
- Number of offers available: 1
- Company/Institute: University of Latvia, House of Science
- Country: Latvia
- City: Riga
- Postal Code: LV-1004
- Street: Jelgavas iela 3
Contact
- City: Riga
- Website: https://www.lu.lv/en/
- https://eztf.lu.lv/en/about-us/faculty-departments/department-of-mathematics/
- Street: Jelgavas iela 3
- Postal Code: LV-1004
- E-Mail: janis.bajars@lu.lv