The Mathematics and Computer Science Division at Argonne National Laboratory seeks self-motivated and independent Postdoctoral researcher to develop and apply state-of-the-art probabilistic machine learning techniques for the development of efficient and robust surrogate models for scientific machine learning applications. This is envisaged to support various scientific domains that are relevant to Argonne's mission, including but not limited to Fusion Science, High Performance Computing, Aerostructure Manufacturing, and High Energy Physics.
Probabilistic machine/deep learning and, especially, the Bayesian framework provides an exciting avenue to address some of the challenges related to reliability and robustness encountered by their deterministic counterparts. However, the Bayesian inference for large models needed for scientific machine learning can be computationally intensive. The selected candidate will be building on the ongoing research at Argonne by developing efficient probabilistic modeling and inference through a combination of novel non-parametric sparsity-inducing Bayesian priors, information-theoretic learning, neural architecture search and hybrid sequential - parallel ensembling, and probabilistic programming techniques.
Argonne is the home of a DOE Leadership Computing Facility which will be home to Aurora, an exascale computing system, and is already equipped with cutting edge machine-learning accelerators such as Cerebras, Sambanova, Groq, Graphcore, Nvidia DGX-A100, and many other advanced AI platforms. The selected candidate will also have the unique opportunity to develop and scale probabilistic machine learning approaches for various DOE scientific domains using these state-of-the-art computing resources.
Position Requirements
- Recent or soon-to-be completed Ph.D. (typically completed within 3 years) in Computer Science or Mathematics with strong background in one or more of the following: Statistical machine learning, Bayesian deep learning, probabilistic and differentiable programming, probability and measure theory.
- Evidence of relevant achievements in probabilistic machine/deep learning, deep latent variable models, uncertainty quantification or Bayesian inference algorithms research and development, as demonstrated with technical publications, presentations, or software releases.
- Strong development skills in deep learning frameworks (eg., PyTorch, TensorFlow, or Jax)
- Familiarity with probabilistic programming frameworks (eg., Tensorflow Probability, Pyro, Gen, Edward2)
- Strong skill in written and oral communication.
Preferred experience:
- Machine learning on high-performance computing systems and AI accelerators.
- Previous scientific machine learning experience.
- Familiar with Information theoretic principles, generative models, and ensembling UQ techniques.
Job Family
Postdoctoral Family
Job Profile
Postdoctoral Appointee
Worker Type
Long-Term (Fixed Term)
Time Type
Full time
As an equal employment opportunity and affirmative action employer, and in accordance with our core values of impact, safety, respect, integrity and teamwork, Argonne National Laboratory is committed to a diverse and inclusive workplace that fosters collaborative scientific discovery and innovation. In support of this commitment, Argonne encourages minorities, women, veterans and individuals with disabilities to apply for employment. Argonne considers all qualified applicants for employment without regard to age, ancestry, citizenship status, color, disability, gender, gender identity, gender expression, genetic information, marital status, national origin, pregnancy, race, religion, sexual orientation, veteran status or any other characteristic protected by law.
Argonne employees, and certain guest researchers and contractors, are subject to particular restrictions related to participation in Foreign Government Sponsored or Affiliated Activities, as defined and detailed in United States Department of Energy Order 486.1A. You will be asked to disclose any such participation in the application phase for review by Argonne's Legal Department.
All Argonne offers of employment are contingent upon a background check that includes an assessment of criminal conviction history conducted on an individualized and case-by-case basis. Please be advised that Argonne positions require upon hire (or may require in the future) for the individual be to obtain a government access authorization that involves additional background check requirements. Failure to obtain or maintain such government access authorization could result in the withdrawal of a job offer or future termination of employment.
#J-18808-Ljbffr
