Machine Learning Physics Graduate Student

LLNL•Livermore, CA

1d•$6,752 - $8,201•Onsite

About The Position

We have multiple openings for Machine Learning Graduate Student Interns to engage in practical research experience to further their educational goals. You will work on multidisciplinary projects, such as development of classical empirical and machine learning interatomic potentials, discovery of partial differential equations (PDEs), numerical solutions of partial differential equations to model material behavior at continuum scale and analysis of large atomic datasets. These positions are in in the Equation of State Materials Theory Group of the Physics Division of the Physical & Life Sciences Directorate. This position requires full-time on-site presence due to the nature of the work.

Requirements

Must be eligible to access the Laboratory in compliance with Section 3112 of the National Defense Authorization Act (NDAA).
Continuing student in good standing at an accredited institution of higher education pursuing a graduate degree in Physics or related field.
Research background with a record of publication.
Experience in writing codes (in C/C++ and Python) and a background in Materials Science/Engineering/Physics/Applied Mathematics.
Excellent skills in written and verbal communication, as well as teamwork.

Nice To Haves

Experience in parallel computing, porting codes to GPUs, experience in numerical solutions of partial differential equations.

Responsibilities

Develop parallel C/C++/Python codes to train, test and evolve (a) PDEs (for phase field and phase field crystal models) discovered from data, and (b) interatomic potentials developed from quantum simulations.
Explore the use of machine learning methods to discover and evolve PDEs for phase field and phase field crystal models.
Analyze results, provide weekly updates and present work at poster sessions
Review literature in the field of study, document results and write papers.
Perform other duties as assigned.