(CLOSED) FY2022 Artificial Intelligence Research for High Energy Physics (DE-FOA-0002705)

Sponsor Name: 
DOE
Description of the Award: 

Applicant institutions are limited to no more than four (4) letters of intent and four (4) applications for single-institution Seed applications. Only DOE/NNSA National Laboratories are eligible to submit multi-institution applications under this FOA. There is no limit to the number of teams that an institution may join as a non-lead team member.

This FOA will invest in AI research applied to the HEP program following previously identified national research priorities and Basic Research Needs workshops conducted by SC.[1, 2, 3] The objectives are to support AI research that extends the scientific reach of existing HEP programs well beyond what is currently achievable including across frontiers and experiments; uses HEP to improve the understanding of the theoretical capabilities and limitations of fundamental AI; or develops shared public HEP datasets and computing environments for AI training and testing. The aim of this FOA is to consolidate existing parallel efforts and to foster new directions towards these objectives. Applications are sought that primarily focus on one of the following topic areas, though research may benefit secondary areas.

AI for HEP – The scientific objectives and priorities for the field recommended by the High Energy Physics Advisory Panel (HEPAP) are detailed in its long-range strategic Particle Physics Project Prioritization Plan (P5).[4] Applications are sought that are well aligned with the HEP program priorities. AI research that advances the P5 science drivers, or development of new AI-based technologies that expand paths of investigation for HEP beyond what was considered in the P5 report are encouraged. Ambitious applications of AI benefitting multiple HEP programs or experiments coherently and which are of broad interest are especially sought, as are innovative applications that can deliver significant advances to HEP experimental reach or theoretical understanding.

HEP for AI – This FOA also seeks to support research that makes use of unique aspects of HEP to improve the understanding of the theoretical capabilities and limitations of fundamental AI. Research into robust scientific ML, data intensive ML, ML-enhanced Modeling and Simulation, Uncertainty Quantification, and Physics Informed ML that exploits HEP theoretical understanding, experimental data, or simulations to provide insight into general AI/ML methods are encouraged. An example of a possible research topic in this area would be evaluation of various Physics Informed ML techniques compared to training more traditional networks.

HEP AI Ecosystem – Proposed work toward production of open datasets, collaboration with industrial or national laboratory partners, as well as development of “ecosystem” software allowing for straightforward training and deployment of models is equally encouraged. Applications that address democratic access among all-sized institutions to computing resources and continued development and retention of the workforce for the products being developed are especially welcome. Examples of possible topics in this area would be curating HEP datasets for public access, or integration of modern ML software into standard HEP tools.

Applications for work currently supported by DOE or other funding entities are outside the scope of this FOA. Applications that are not in direct support of HEP research (e.g., conferences, experimental operations, conceptual research and development (R&D), design, or fabrication directed towards a specific project, etc.) may not be submitted in response to this FOA. They may be submitted to the annual SC FOA or to other applicable SC FOAs published at https://www.grants.gov/.

Scientific and Technical Areas of Interest

The following program descriptions are offered to provide more in-depth information on scientific and technical areas of interest to HEP:

Program Website: https://science.osti.gov/hep/.
The mission of the HEP program is to understand how the universe works at its most fundamental level, which is done by discovering the elementary constituents of matter and energy, probing the interactions between them, and exploring the basic nature of space and time.

The Energy Frontier - where powerful accelerators are used to create new particles, reveal their interactions, and investigate fundamental forces;

The Intensity Frontier - where intense particle beams and highly sensitive detectors are used to pursue alternate pathways to investigate fundamental forces and particle interactions by studying events that occur rarely in nature, and to provide precision measurements of these phenomena;

The Cosmic Frontier - where non-accelerator-based experiments observe the cosmos and detect cosmic particles, making measurements of natural phenomena that can provide information about the nature of cosmic acceleration, including dark energy and the cosmic microwave background, search for dark matter particles, and studying properties of the universe that impact our understanding of matter and energy;

Theoretical High Energy Physics, where the vision and mathematical framework for understanding and extending the knowledge of particles, forces, space-time, and the universe are developed;

Accelerator Science and Technology Research and Development, where the technologies and basic science needed to design, build, and operate the accelerator facilities essential for making new discoveries are developed;

Detector Research and Development, where the basic science and technologies needed to design and build the High Energy Physics detectors essential for making new discoveries are developed.

Sponsor LOI Deadline: 
Apr 21, 2022
Sponsor Final Deadline: 
May 24, 2022
OSVPR Application or NOI Instructions: 

Interested applicants should upload the following documents in sequence in one PDF file (File name: Last name_DE-FOA-0002705_2022) no later than 4:00 p.m. on the internal submission deadline:

1. Cover Letter (1 page, pdf):

  • Descriptive title of proposed activity
  • PI name, departmental affiliations(s) and contact information
  • Co-PI's names and departmental affiliation(s)
  • Names of other key personnel
  • Participating institution(s)
  • Number and title of this funding opportunity

2. Project Description (no more than two pages, pdf) identifying the project scope that addresses the key aspects and elements of the sponsor's solicitation, principal investigators, collaborators, and partner organizations. Figures and references, if included, must fit within the two page limit.

3. 2-page CV's of Investigators

Formatting Guidelines:

Font/size: no smaller than 11 pt.
Document margins: 1.0” (top, bottom, left and right)
Standard paper size (8 ½” x 11)

To be considered as a Penn State institutional nominee, please submit a notice of intent by the date provided directly below.
This limited submission is in downselect: 
Penn State may only submit a specific number of proposals to this funding opportunity. The number of NOIs received require that an internal competition take place, thus, a downselect process has commenced. No Penn State researchers may apply to this opportunity outside of this downselect process. To apply for this limited submission, please use this link:
OSVPR Downselect Deadline: 
Tuesday, April 12, 2022 - 4:00pm
For help or questions: 

Questions concerning the limited submissions process may be submitted to limitedsubs@psu.edu.

Notes: 
No Applicants, Now first come, first served - Contact LimitedSubs@psu.edu if you wish to apply