Research associate (postdoc) - Large-Scale Bayesian Uncertainty Quantification in Computational Biomechanics : Luxembourg, Luxembourg

The University of Luxembourg has the following vacancy within the Unit Engineering of its Faculty of Science, Technology and Communication (FSTC).

Research associate (postdoc) – Large-Scale Bayesian Uncertainty Quantification in Computational Biomechanics (M/F)

· Ref: F1R-ING-PEU-14RLTC
· 1 year fixed-term contract, full-time (40 hrs/week), may be extended
· Employee status

The computational mechanics group of Prof. Stephane Bordas is looking for a postdoctoral researcher to work on Large-Scale Bayesian Uncertainty Quantification problems in Computational Biomechanics. The successful candidate will join a dynamic team of computational scientists, mathematicians and engineers in the Research Unit in Engineering Science at the University of Luxembourg.

Mission:
The candidate will be responsible for building on preliminary work on quantifying the uncertainty in the recovered parameters of a geometrically non-linear hyperelastic soft-tissue model.

We want to scope the project direction with the successful candidates' interests in mind. Possible directions include, but are not limited to: 
· Hamiltonian or other efficient Monte Carlo methods,
· dynamic data-driven model updating,
· model order reduction techniques for Monte Carlo sampling,
· scalable maximum a posteriori (MAP) estimators,
· variational adaptivity for Bayesian inverse problems,
· preconditioning and solution strategies for large-scale hyperelastic inverse problems,
· efficient eigenvalue solvers and low-rank update algorithms.

Profile:
Candidates with a PhD in applied mathematics, computational mechanics, scientific computing, engineering or statistics are encouraged to apply. Experience in one or more of the following areas would be ideal: 
· PDE-constrained optimisation,
· Monte Carlo methods,
· error estimation,
· Bayesian statistics,
· complex hyperelastic constitutive models,
· model order reduction,
· solving PDEs with FEniCS (http://fenicsproject.org),
· solving PDE-constrained optimisation problems with dolfin-adjoint (http://dolfin-adjoint.org),
· large scale linear algebra with PETSc (http://www.mcs.anl.gov/petsc/).