University of Latvia
Closing soon
2 weeks ago
PhD position in Neural Operators for Physical Systems University of Latvia in Latvia
Degree Level
PhD
Field of study
Computer Science
Funding
Available
Deadline
Oct 31, 2026
Country
Latvia
University
University of Latvia
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.
More information can be found here
Keywords
Suggested positions
About this position
The University of Latvia is offering a funded PhD position in the Department of Mathematics for research in physics-informed neural operators for modelling complex physical processes governed by partial differential equations.
The project sits at the intersection of applied mathematics, deep learning, scientific computing, and mathematical physics. It focuses on operator learning methods such as DeepONet and Fourier Neural Operators, including physics-informed extensions. The aim is to deepen theoretical understanding and develop practical improvements that make neural operators more robust, better generalising, and more physically consistent as fast surrogate models for PDE-based simulations.
Research applications highlighted in the advert include hydrodynamics, especially crystal growth processes and wave dynamics in coastal and harbour environments. The work is theory-driven but closely tied to computational applications, making it attractive to candidates who want to combine rigorous mathematical analysis with modern machine learning.
The University states that it is particularly interested in applicants with a background in applied mathematics, especially those working at the interface of mathematical modelling or theoretical physics. A strong understanding of PDEs and numerical methods is expected, together with some prior exposure to deep learning and modern ML frameworks.
The position is funded for up to three years. The project-funded salary is approximately 2000 EUR per month before tax, for the funded workload within the project rather than a full-time appointment. The advert also notes possible 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.
Eligibility highlights: Master’s degree in applied mathematics, mathematical physics, computational science, or a related field; strong PDE and numerical methods background; experience in scientific computing and modelling; familiarity with deep learning and frameworks such as PyTorch, TensorFlow, or JAX; excellent English; and the ability to work independently.
Application deadline: 31 October 2026, 23:59 (Europe/Riga).
Contact/apply: Jānis Bajārs, [email protected].
Institution: University of Latvia, House of Science, Riga, Latvia.
Funding details
Available
What's required
A Master’s degree in applied mathematics, mathematical physics, computational science, or a related field is required. Applicants should have 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 such as PyTorch, TensorFlow, or JAX, strong programming and analytical skills, excellent English communication skills, and the ability to conduct independent research. Knowledge or experience in physics-informed machine learning, neural operators such as DeepONet or Fourier Neural Operators, hydrodynamics, wave dynamics, or related areas is an advantage.
How to apply
Apply by email to [email protected]. Use the subject line shown in the posting and include the requested application materials in your message. For details, consult the University of Latvia Department of Mathematics contact information provided in the listing.
Ask ApplyKite AI
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.