Share this Job

Research Scientist - Causal Machine Learning and Materials Simulations

Date: Sep 15, 2022

Location: Oak Ridge, TN, US, 37830

Company: Oak Ridge National Laboratory

Requisition Id 8649 

Overview: 

The Computational Chemistry and Nanomaterials Sciences Group at Oak Ridge National Laboratory is seeking a highly motivated scientist to deliver groundbreaking computational chemical and materials science at the forefront of the field through the development and use of machine learning, deep learning, multiscale simulations and high-performance computing (HPC). This group leads and contributes to the development of Quantum Monte Carlo, First Principles, and Hubbard electronic structure methods, reactive atomic-scale simulation methods including empirical force fields and approximate density functional theory, and machine learning methods for simulation and experimental data. Examples of scalable codes for HPC platforms currently developed include LAMMPS and DFTB++, among others.

Specific areas of interest include:

  • Causal models, machine learning, deep learning, symbolic approaches via statistical learning to interventional models relying on concepts of causality.
  • Atomistic simulation methods coupled with causal representation learning and group theory for functional, quantum materials.
  • Understand fundamental physical mechanisms operating in functional materials via multimode coupling.
  • Common-cause, causal chain and common-effect models with interventions, planning and reasoning, specific to tune material functionalities as well as design novel materials.
  • Use of these tools to investigate underlying structure-property relationships beyond standard built-in correlations, build frameworks bridging experiments and theory.
  • Strong collaborations with experimental researchers at ORNL in the area of microscopy experiments and synthesis.

 

This position will be part of the Advanced Computing Methods for Physical Sciences Section within the Computational Sciences & Engineering Division (CSE). The Advanced Computing Methods for Physical Sciences Section delivers multiscale, multifidelity computational models and systems developing algorithms and analytics for physical sciences.

 

CSE focuses on transdisciplinary computational science and analytics at scale to enable scientific discovery across the physical sciences, engineered systems, and biomedicine and health sciences. It provides foundations and advances in quantum computation and information science and develops community applications, data assets, and technologies to advance crosscutting science outcomes.

 

Major Duties/Responsibilities:

We are seeking a candidate with expertise in their area of technical expertise that will initiate and perform independent R&D on an ongoing basis.

Responsibilities for this position include working closely with the project team to create, evaluate, and publish novel research ideas, supervise and mentor students, and collaborate with our partners and sponsors. Research activities include, but are not limited to:

  • Development and evaluation of statistical and causal machine learning tools for designing and understanding  functional materials using the leadership class high performance computing facilities available at ORNL and other DOE facilities
  • Publishing papers in high-quality refereed journals and conferences such as journals from theA the American Physical Society, American Chemical Society, Nature, etc.
  • Actively collaborating with industry, academia, government labs, and applications developers in a variety of venues.
  • Contributing to research proposals and reporting to major funding agencies. 

 

Basic Qualifications Required:

  • Requires a Ph.D. in computational condensed matter physics, chemistry, civil engineering, computer science, or a related discipline, and at least 2 years of relevant research experience, beyond the PhD.
  • A well established track record of research in an area relevant to one or more areas of expertise of the group.
  • Basic knowledge of reactive atomistic modeling techniques for simulations of materials.

 

Preferred Qualifications:

  • Programming experience in atomistic simulation codes (e.g. VASP, Quantum Espresso, ASE, LAMMPS, DFTB++).
  • Programming experience in causal machine learning, probabilistic programming, statistical methods (e.g. Python, PyTorch, TensorFlow, JAX, Julia).
  • Familiarity with MatMiner, RDKit, PaDEL or other materials informatics packages.
  • Demonstrated experience in probabilistic and causal modeling for materials to understand structure-property relationships, solid structure and functionality.
  • The ability to take initiative on research insights to bring them to fruition through publication or demonstration on mission applications.
  • Excellent interpersonal skills, oral and written communication skills, and strong personal motivation.
  • Ability to function well in a fast-paced research environment, set priorities to accomplish multiple tasks within deadlines, and adapt to ever changing needs. 

 

Benefits at ORNL:

UT Battelle offers an exceptional benefits package to include matching 401K, Pension Plan, Paid Vacation and Medical / Dental plan. Onsite amenities include Credit Union, Medical Clinic, and free Fitness facilities.   

 

This position will remain open for a minimum of 5 days after which it will close when a qualified candidate is identified and/or hired.

We accept Word (.doc, .docx), Adobe (unsecured .pdf), Rich Text Format (.rtf), and HTML (.htm, .html) up to 5MB in size. Resumes from third party vendors will not be accepted; these resumes will be deleted and the candidates submitted will not be considered for employment.


If you have trouble applying for a position, please email ORNLRecruiting@ornl.gov.


ORNL is an equal opportunity employer. All qualified applicants, including individuals with disabilities and protected veterans, are encouraged to apply.  UT-Battelle is an E-Verify employer.


Nearest Major Market: Knoxville