Postdoc: Multiscale thermodynamics, superfluids, and machine learning
This postdoc position is about geometric reduction techniques in non-equilibrium thermodynamics and
statistical physics, and about applications in superfluids.
Reduction techniques are important because they find low-dimensional patterns in otherwise exceedingly complex
behavior.
The reduction techniques are then applied to models and simulations of superfluid helium-4, where many
theoretical problems remain open (like the energy of a compressible superfluid, quantum vortices, etc). We
focus on geometric technique
s (Hamiltonian mechanics, contact geometry, etc), so that the results are independent of representation.
The position consists of three parts, each of which can become the main topic:
- Numerics: solving hyperbolic partial differential equations for superfluids, smoothed particle
hydrodynamics, or Hamilton-Jacobi equations.
- Multiscale non-equilibrium thermodynamics and statistical physics.
- Machine learning for recognition of patterns in statistical physics or in data describing superfluid
helium.
Concrete objectives of the research (to be carried out in collaboration with me, my colleagues, and students)
will be set upon personal agreement, based on the PostDoc's experience and interest.
- Condition: Ph.D. degree obtained after Dec 31, 2014.
- The salary will be CZK 52 K/month (plus the social and health insurance, can vary upon agreement).
- The position is located in Prague, Czech Republic (Mathematical Institute, Faculty of Mathematics and
Physics, Charles University).
- The position is available from July 2023 (or later) to December 2025 (may be shorter if needed).
- Contact: Michal Pavelka, pavelka@karlin.mff.cuni.cz, www.karlin.mff.cuni.cz/~pavelka
|