Slots: 2 slots per research area specified in Section III. One slot taken for Topic E.
Deadlines
Internal Deadline: Friday, February 7th, 2025, 5pm PT
LOI: February 25th, 2025
External Deadline: May 6, 2025
Award Information
Award Type: Grant
Estimated Number of Awards: 20
Anticipated Award Amount: $200,000 – $1,000,000
Who May Serve as PI: Standard DOE language.
Link to Award: https://www.grants.gov/search-results-detail/358344
Process for Limited Submissions
PIs must submit their application as a Limited Submission through the Research Initiatives and Infrastructure (RII) Application Portal: https://rii.usc.edu/oor-portal/. Use the template provided here: RII Limited Submission Applicant Template
Materials to submit include:
- (1) Two-Page Proposal Summary (1” margins; single-spaced; standard font type, e.g. Arial, Helvetica, Times New Roman, or Georgia typeface; font size: 11 pt). Page limit includes references and illustrations. Pages that exceed the 2-page limit will be excluded from review. You must use the template linked above.
- (2) CV – (5 pages maximum)
Note: The portal requires information about the PIs in addition to department and contact information, including the 10-digit USC ID#. Please have this material prepared before beginning this application.
Purpose
The DOE SC program in Advanced Scientific Computing Research (ASCR) hereby announces its interest in basic research to explore potentially high-impact approaches in scientific computing and extreme-scale science.
RESEARCH OPPORTUNITIES
Exploratory Research for Extreme-Scale Science (EXPRESS) opportunities exist for the following research topics:
A) Quantum computation based on topological concepts B) Local and Campus-Area Quantum Networking for Next Generation Parallel Quantum
Computing
C) Neuromorphic Computing
D) Computational Physical Systems
E) Deep Understanding of AI Models
Applications submitted in response to this NOFO must substantially address one among the preceding list of research topics. Additional details about each topic are provided below.
A) Quantum computation based on topological concepts
Technical Contact/Program Manager: Marco Fornari, Marco.Fornari@science.doe.gov
The ASCR Workshop on Basic Research Needs in Quantum Computing and Networking [1] highlights the importance of enhancing the resilience of quantum systems to noise and errors suggesting that co-design techniques could contribute to error prevention, protection, mitigation, and correction. Topology studies the features of mathematical objects that are preserved through continuous transformations. In quantum science, uses of topology include understanding small perturbations affecting quantum states and special topological properties that can potentially protect quantum states from decay and decoherence. As a result, quantum states with specific topological characteristics lead to promising advances in qubit technologies, quantum memories, and quantum algorithms.
The principles of topological quantum computing have been extensively studied; practical implementations are, however, still in their infancy due to the difficulty of stabilizing and controlling anyons in physical systems. Nevertheless, algorithmic developments guided by topological considerations have been quite successful in developing efficient techniques to control quantum errors. In addition, topological quantum systems have motivated many applications on NISQ hardware to simulate effects and test theoretical conjectures.
Research Area
This topic seeks to exploit topological concepts and strategies to advance models and algorithms toward the possible realization of universal quantum computing and quantum memories, including the integration of topological principles in current quantum computing approaches.
Basic research on ways to potentially exploit topological quantum computing, advancements aimed to expand on algorithms for topologically protecting quantum states, and proposals focused on designing quantum topological models relevant to quantum computation are central to this topic. Models and algorithms must be accompanied by sound theoretical and computational analysis, possibly including verification and validation strategies.
Out of Scope
Out of scope are pre-applications and applications that:
- Fail to address the research area specified above; or
- Research quantum security protocols; or
- Focus on quantum simulation of topological models without addressing computing aspects; or
- Solely focus on quantum materials; or
- Propose to physically construct qubits.
Pre-applications and applications that meet the out-of-scope criteria described above will be declined without review.
B) Local and Campus-Area Quantum Networking for Next Generation Parallel Quantum Computing
Technical Contact: Kalyan Perumalla, Kalyan.Perumalla@science.doe.gov
Quantum networking involves effective communication of quantum information among geographically distributed quantum systems, separated by short or long distances. Sources of quantum information in the network communication include qubits used in different forms of quantum computing or output from various quantum sensing devices.
This topic seeks advancements in quantum networking over short distances to enable parallel quantum computing within a building and integration of quantum information sources to storage and quantum computing across a campus area. The interconnection of different co-located quantum computing systems is aimed at increasing the scale of quantum computation (e.g., aggregate number of qubits) and at progressing towards an architecture of flexible connectivity of quantum devices across a laboratory or university campus, potentially composing heterogeneous quantum computing hardware, including broadening of networking from exchange of physical qubits to logical qubits.
Research Area
The specific aim of this topic is to support quantum science needed to effectively scale quantum computing and enable flexible exchanges of coherent quantum information via locally networked heterogeneous quantum systems. Proposals must address one or more research advancements in the aforementioned directions in local and campus-area quantum networking. Research must be aimed at advancing our understanding of aspects, such as core concepts, devices, architectures, integration, and interfaces, that are necessary for a quantum counterpart to the current infrastructures of classical local and campus area networks within the scientific and other facilities.
Out of Scope
Topics that are out of scope include:
- Research solely focused on quantum materials
- Research involving wide-area quantum repeaters
- Research involving wide-area quantum networking architectures and testbeds
- Research primarily focused on coexistence of quantum and classical communication
- Research in quantum security protocols.
Projects that fall within the out-of-scope areas described above will be declined without review.
References
ASCR Basic Research Needs in Quantum Computing and Networking” (Report), July 2023, https://doi.org/10.2172/2001045.
C) Neuromorphic Computing
Technical Contact: Robinson Pino, Robinson.Pino@science.doe.gov
Neuromorphic computing spans a broad range of scientific disciplines from emerging devices, engineering, computer science to neuroscience, all of which are required to solve the neuromorphic computing grand challenge. The aim of this topic is to support high risk and high reward research and development of computational neuroscience and biologically inspired computing architectures with emphasis on analog applications on energy-efficient emerging microelectronic circuitry. The aim is to investigate and prototype foundational neuromorphic computing architecture primitive circuits that will replicate neurobiological processes and accelerate AI for scientific research applications while delivering energy efficiency.
Research Area
This topic is specific to the modeling, fabrication, and prototyping of neuromorphic computing circuit primitives for generalizable applications emphasizing analog approaches, emerging devices, algorithms, and energy efficiency. Proposals must address state-of-the-art and propose to investigate basic research approaches beyond the state-of-the-art. In addition, pre-proposals and proposals must clearly articulate a computational neuroscience-based justification (beyond high level descriptions or representations of neuron and synapses), circuit primitive, and rationale for the proposed research approach through energy-efficient hardware implementations that leverage emerging devices and CMOS. This research topic focuses on demonstrating biologically realistic neuromorphic circuits primitives that capture the functionality of neural systems found in nature.
Out of Scope
Topics that are out of scope include:
- Research that does not address the specific topic described above;
- Statistical, probabilistic, reservoir, crossbar, hyper dimensional, mix-signal, cryogenic, quantum, or digital approaches to neuromorphic computing;
- Projects that solely focus on materials-science research or software;
- Projects that do not propose fabrication;
- Projects that do not propose prototyping of emerging devices (non-CMOS) and CMOS.
- Projects that fall within the out-of-scope areas described above will be declined without review.
References
DOE Workshop: Neuromorphic Computing for Science, September 2024, https://www.orau.gov/2024NeuromorphicComputing.
D) Computational Physical Systems
Technical Contact: Hal Finkel, Hal.Finkel@science.doe.gov
Analog computing, which involves computation using continuous quantities, has been an essential aspect computing technology from the earliest eras to the present day. However, unlike the regimes of digital computing, which involves computing using discrete quantities and operations, both the theory of, and techniques for, analog computing remain underdeveloped.
Fortunately, work over the past decade has opened new avenues for the development of computer-science and mathematical techniques that are applicable to the design of complex chemical and physical systems, with applications ranging from the construction of data storage and computing systems to synthetic biology and nanotechnology. Thus, progress in analog computing might not only prove essential to the creation of future energy-efficient computing technologies, but can also drive innovation across many other domains that aim to create complex physical devices and systems. In the summer of 2024, the research community and other stakeholders gathered for the ASCR Analog Computing for Science Workshop [1] to explore recent advancements and future research opportunities.
Research Area
Each pre-application and application must propose fundamental research that aims to advance our understanding of how complex physical systems can be created whose evolution in time can be characterized as analog computation. The specific focus of the proposed research should include exploring the mathematical frameworks, programming paradigms, or creation of novel devices needed to design, co-design, and/or enable programmatic use of, analog computing systems, especially at large scale. The proposed work must include an aim to characterize the properties of the systems needed to support articulated relevant classes of algorithms and programming constructs. The techniques investigated should be scalable to use cases of non-trivial complexity, account for the inherent parallelism of the relevant physical systems, and account for the constraints implied by the targeted class of physical systems while optimizing relevant objectives which may include, but are not limited to: minimizing the number of dependent variables, maximizing the execution speed, minimizing the execution energy, or maximizing resilience.
Out of Scope
Out of scope are pre-applications and applications that:
- Fail to address the research area specified above; or
- Investigate materials physics without a clear focus on novel devices for analog computation; or
- Fail to describe the class of physical systems applicable to the techniques under investigation; or
- Propose work applicable only to a single application domain.
Pre-applications and applications that meet the out-of-scope criteria described above will be declined without review.
References
2024 Analog Computing for Science Workshop. See the pre-workshop document, accepted position papers, and other posted materials.
https://www.orau.gov/2024AnalogComputingWorkshop.
E) Deep Understanding of AI Models
Technical Contact: Steve Lee, Steven.Lee@science.doe.gov
AI has a critical and growing impact on nearly all aspects of modern science [1], and advancement in AI has been driven by both significant advancement in practical techniques for the creation and use of large-scale AI models and by a solidifying, but still relatively nascent, theoretical understanding of the AI training process and the models that emerge from that process. To enable the creation and use of AI models for science, many aspects of the AI training and inference processes require optimization, including speed, energy efficiency, memory and data usage, and accuracy of the predicted values and the characterized uncertainty of those values.
Research Area
Each pre-application and application must propose computer-science or applied-mathematics research that aims to advance our understanding of the AI-model training process and the representations that are developed by the training process within the resulting models. Of particular interest is the study of training dynamics [2], which views the AI model-training process as the evolution of a dynamical system, and allows building on the decades of advancement in applied mathematics for the evolution, characterization, and optimization of dynamical systems to improve the process of training large-scale AI models. Of particular interest is also the study of the representations of scientific concepts developed within AI models, including the generalizability and stability of those representations, the emergence of in-context learning and other reasoning capabilities, grokking [3] and other potential phase-
transition phenomenon, and the interplay between those representations and properties of AI training algorithms.
This research area includes both the study of scientifically relevant aspects of large-language models, models that are trained on non-language scientific data, and models that are trained using a combination of both language and non-language scientific data.
Out of Scope
Out of scope are pre-applications and applications that:
- Fail to address the research area specified above; or
- Fail to focus on fundamental advances in AI for science; or
- Focus on purely empirical studies of AI models or their scaling properties; or
- Do not have computer-science or applied mathematics research as the primary focus of the proposed work; or
- Provide discipline-specific and/or application-specific solutions that do not generalize to multiple applications; or
- Focus on the development of applications or approaches for quantum computers.
Pre-applications and applications that meet the out-of-scope criteria described above will be declined without review.
Program Goals, Objectives, and Priorities
The Office of Science’s (SC) mission is to deliver scientific discoveries and major scientific tools to transform our understanding of nature and advance the energy, economic, and national security of the United States (U.S.). SC is the Nation s largest Federal sponsor of basic research in the physical sciences and the lead Federal agency supporting fundamental scientific research for our Nation s energy future.
SC accomplishes its mission and advances national goals by supporting:
- The frontiers of science-exploring nature s mysteries from the study of fundamental subatomic particles, atoms, and molecules that are the building blocks of the materials of our universe and everything in it to the DNA, proteins, and cells that are the building blocks of life. Each of the programs in SC supports research probing the most fundamental disciplinary questions.
- The 21st Century tools of science providing the nation s researchers with 28 state-of-the-art national scientific user facilities, the most advanced tools of modern science, propelling the U.S. to the forefront of science, technology development, and deployment through innovation.
- Science for energy and the environment paving the knowledge foundation to spur discoveries and innovations for advancing the Department s mission in energy and environment. SC supports a wide range of funding modalities from single principal investigators to large team-based activities to engage in fundamental research on energy production, conversion, storage, transmission, and use, and on our understanding of the earth systems.
SC is an established leader of the U.S. scientific discovery and innovation enterprise. Over the decades, SC investments and accomplishments in basic research and enabling research capabilities have provided the foundations for new technologies, businesses, and industries, making significant contributions to our nation s economy, national security, and quality of life.
Visit our Institutionally Limited Submission webpage for more updates and other announcements.