Oak Ridge National Laboratory, Applied Math Group

Position Title: Staff Mathematician: Data Analytics and Machine Learning
Position Location: Oak Ridge, Tennessee 37831, United States [map]
Application Deadline: 2018/11/31 finished (2018/09/06, finished 2019/06/02, listed until 2018/11/31)
Position Description:    

*** this position has been closed, and no new applications will be accepted. ***

The Computational and Applied Mathematics (CAM) Group is seeking an exceptional mathematician to work on the development of rigorous theory and algorithms for applying machine learning (ML) techniques, customized to large-scale scientific application interests of the Department of Energy (DOE). In particular, the application of ML to scientific discovery will require an entirely new focus on algorithmic development and mathematical rigor, than what has been produced by the commercial sector. The application of ML to problems of interest to the DOE, or for scientific discovery, must leverage the large body of high performance algorithmic development with mathematics driving advances in the following major thrusts of research: (1) formulate novel mathematical tasks which are achievable given the type (plentiful or sparse) and form (labeled or unlabeled, static or online, deterministic or noisy) of observational data; (2) design of architectures/models which accurately capture the complexities of the data, with robust estimates of confidence in predictions on defined domains; (3) fast and scalable algorithms to fit the proposed models to data, with a theory that explains the convergence and success of these techniques; and (4) detailed re-analysis of the performance of the trained models to determine underlying processes that govern the given data.

This particular position focuses on the potential for the use and development of tools from the mathematical sciences, as opposed to traditional statistical theory or engineering tools, to support and advance the understanding of deep learning and/or ML technology, and to accelerate algorithmic performance on leadership computing platforms.

The position requires collaboration within a multi-disciplinary research environment consisting of mathematicians, computational scientists, computer scientists, experimentalists, and engineers/physicists conducting basic and applied research in support of the Laboratory’s missions.

Specific responsibilities include: • Participating in the development of innovative theoretical analysis and computational methodologies for deep and/or machine learning technologies, customized to large-scale scientific applications, including: additive manufacturing; quantum material design; scientific imaging for material discovery; medical and tomographic imaging; population and evolutionary dynamics; cyber security and power grid analysis; consumer behavior and market analysis; as well as uncertainty quantification for engineered systems. • Collaboration with experts from various scientific disciplines and applications, and following team planning, goals and quality processes. • Authoring peer reviewed papers, technical papers, reports and proposals. • Maintaining memberships in professional, academic, and research organizations.

Basic Qualifications: • Ph.D. in Mathematics, Engineering, or Data/Computational Science. • 2+ years of relevant research experience outside of Ph.D.

Preferred Qualifications: • Strong theoretical background in: constructive approximation; functional, complex, and real analysis; as well as numerical analysis and approximation of partial differential equations. • Expertise in deep learning and reinforced learning; graphical/network models; kernel methods; sparse grid methods; compressed sensing, thresholding, and greedy approaches; stochastic optimization; as well as convex and non-convex regularizations. • Demonstrated experience in the design and implementation of numerical algorithms in one or more high-level computing languages within a team environment. • Ability to program in Python and knowledge of Machine Learning/Deep Learning frameworks, such as TensorFlow, PyTorch, etc. • Effective interpersonal skills. • Demonstrated written and oral communication skills.

Compensation Band Level: Associate R&D Staff Member, Band 2/R&D Staff Member, Band 3

Application Process: Apply to the job posting through the ORNL Career’s Website, and include: • A complete curriculum vitae; • A rigorous Statement of Research, that provides the significance of you current research as well as the technical background; • Three Letters of Reference, describing the applicants distinct contributions to their field of research. Have your letters sent to Kasi Arnold at arnoldkl@ornl.gov. Use "Staff Mathematician” as the subject line.

CONTACT: For more information about the Householder Fellowship or for technical questions please contact Clayton Webster (webstercg@ornl.gov).

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

Application Materials Required:
Submit the following items online at this website to complete your application:
And anything else requested in the position description.

Further Info:
Oak Ridge National Laboratory, Bethel Valley Rd. MS6164, Oak Ridge, TN 37831

© 2019 MathJobs.Org, American Mathematical Society. All Rights Reserved.