DOE Scientific Machine Learning for Complex Systems

Below is a summary assembled by the Research & Innovation Office (RIO). Please see the full solicitation for complete information about the funding opportunity.

Program Summary

The DOE SC program in Advanced Scientific Computing Research (ASCR) hereby announces its interest in research applications to explore potentially high-impact approaches in the development and use of scientific machine learning (SciML) and artificial intelligence (AI) in the predictive modeling, simulation and analysis of complex systems and processes.

Topic: Uncertainty Quantification for Scientific Machine Learning Modeling and Simulations

Uncertainty quantification (UQ) in scientific computing research is increasingly important for keeping pace with the ongoing advances in the modeling, simulation, and analysis of complex physical systems and processes. A familiar adage is “All models are wrong, but some are useful.” Predictive computational models and simulations differ from real-world experiments in fundamentally different ways and for various reasons. For simulations, complex problems of interest must first be described and formulated in terms of the relevant physical principles, initial and boundary conditions, parameter values, experimental and observational data, and other inputs. These models must be properly discretized, implemented, and programmed for accurate and efficient simulations on high-performance computing systems. However, model inputs are typically known with a limited degree of accuracy or may be inadequately characterized. Another source of real-world and model discrepancy stems from the use of heuristics or approximations from numerical algorithms within the simulation. Furthermore, to avoid the formulation of models that are computationally intractable to simulate, certain complex system phenomena may be ignored or represented with sub-models that are too simplified and/or not accurate enough.

The focus of this funding opportunity is on basic research and development at the intersection of uncertainty quantification (UQ) and scientific machine learning (SciML) applied to the modeling and simulation of complex systems and processes. PRD #5 is the subject of this topic: Machine learning-enhanced modeling and simulation for predictive scientific computing. Scientific computing within the DOE traditionally has been dominated by complex, resource-intensive numerical simulations. However, the rise of data-driven SciML models and algorithms provides new opportunities. Traditional scientific computing forward simulations often are referred to as “inner loop” modeling. The combination of traditional scientific computing expertise and machine learning-based adaptivity and acceleration has the potential to increase the performance and throughput of inner-loop modeling. Such hybrid modeling and simulation approaches offer the opportunity, for example, to combine the versatility of neural networks for function and operator approximations, the domain-knowledge and interpretability of differential equations and operators, and the robustness of high-performance scientific computing software across these areas.

Applications submitted in response to this FOA must substantially address the research topic area and the following three facets in advanced scientific computing:

  1. : What are the most significant or compelling scientific or technical challenges that are driving the development of the proposed approaches?
  2. : In what ways will the proposed research provide new and/or significant enabling technology for scientific computing? What are the potential merits and limitations, particularly with respect to current and emerging high-end computing architectures and ecosystems?
  3. : What is a relevant set of non-trivial metrics for assessing the accuracy and effectiveness of the proposed approaches?


CU Internal Deadline: 11:59pm MST February 13, 2023

DOE Pre-Application Deadline: 3pm MST March 1, 2023 

DOE Application Deadline: 9:59pm MST April 12, 2023

Internal Application Requirements (all in PDF format)

  • Project Summary (1 page maximum): Provide the name of the applicant, the project title, the PI and the PI’s institutional affiliation, any coinvestigators (including any unfunded collaborators) and their institutional affiliations, the objectives of the project, a description of the project, including methods to be employed, and the potential impact of the project (i.e., benefits, outcomes).
  • Lead PI Curriculum Vitae
  • Budget Overview (1 page maximum): A basic budget outlining project costs is sufficient; detailed OCG budgets are not required.

To access the online application, visit:


No more than two pre-applications for each PI at the applicant institution. The PI on a pre-application may also be listed as a senior or key personnel on separate submissions without limitation.

Limited Submission Guidelines

Applicant institutions are limited to no more than four pre-applications as the lead institution in a single- or multi-institutional team.

Award Information

Ceiling: Approximately $1,200,000 per year.

Floor: Approximately $300,000 per year.

Period of Performance: 4 years