Reduced-Order Modeling For Molecular Dynamics

Universities and Institutes of France
December 01, 2023
Contact:N/A
Offerd Salary:Negotiation
Location:N/A
Working address:N/A
Contract Type:Other
Working Time:Full time
Working type:N/A
Ref info:N/A

7 Nov 2023

Job Information

Organisation/Company

Ecole des Ponts

Department

CERMICS

Research Field

Mathematics » Applied mathematics

Researcher Profile

First Stage Researcher (R1)

Country

France

Application Deadline

1 Dec 2023 - 01:00 (Europe/Paris)

Type of Contract

Temporary

Job Status

Full-time

Is the job funded through the EU Research Framework Programme?

H2020 / ERC

Reference Number

101077204

Is the Job related to staff position within a Research Infrastructure?

No

Offer Description

The aim of the present project is to build efficient reduced-order models for parametrized Stochastic Differential Equations, based on dynamical low-rank approximations. Such a task is extremely useful for instance in Molecular Dynamics simulation to assess the influence of parameters defining the interaction potential between the atoms of a given molecular system on the macroscopic behaviour of the total system. Such reduced-order models can then be used as a key ingredient for variance reduction methods in order to accelerate the computation of macroscopic quantities which are defined as averages of observables depending on the trajectories of parameter-dependent stochastic processes. Such dynamical low-rank approximations are classically built using the so-called Dirac-Frenkel variational principle. The latter cannot be used to build accurate approximations of parametrized SDEs on long time scales though. Indeed, the error between the exact trajectories and their approximations grows exponentially with time. The idea of the project is to study theoretically and numerically alternative variational principles for the construction of reduced-order models that lead to new approximations for which the error is much smaller than with the Dirac-Frenkel approach for long time scales.

Requirements

Research Field

Mathematics » Applied mathematics

Education Level

PhD or equivalent
Additional Information Work Location(s)

Number of offers available

1

Company/Institute

Ecole des Ponts ParisTech

Country

France

Geofield

Where to apply

E-mail

[email protected]

Contact

City

MARNE LA VALLEE Cedex 2

Website

https:// www. enpc.fr

Street

6 et 8, avenue Blaise PascalCité DescartesChamps sur Marne

STATUS: EXPIRED

From this employer

Recent blogs

Recent news