Project Grant 2202310
- The National Science Foundation awarded $194,111 under the Engineering federal grant program (CFDA 47.041) to Georgia Tech Research Corporation from August 2022 through July 2025. The funding supports research to develop machine learning-assisted modeling and design of approximate computing techniques to significantly reduce energy consumption in computation-intensive applications such as video/image processing and machine learning. Key products will include input-aware error models for...
- The National Science Foundation (NSF) awarded a $340,000 Project Grant under the Computer and Information Science and Engineering (CISE) program to the University of Missouri System to develop sustainability-aware design methods for approximate deep neural network (DNN) accelerators. The key objectives of this 3-year project are: Investigating efficient neural architecture search techniques for reliable and sustainable approximate DNN accelerator designs. Developing fault mitigation and...
- The National Science Foundation awarded a $1,247,506 Project Grant to the University of Texas at Dallas under the Computer and Information Science and Engineering federal grant program (CFDA 47.070) for the period of October 1, 2022 through September 30, 2025. The grant funds research to comprehend and mitigate errors in analog implementations of on-die neural networks. Specifically, the university will investigate and develop methods to address the impact of manufacturing and operational...
- This Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program to the University of Texas at Dallas (UTD) aims to develop a new sustainability-aware design flow for approximate deep neural network (DNN) accelerators. The $259,741 award seeks to address the energy consumption and fault vulnerability issues of high-precision DNN accelerators used in safety-critical applications. Key objectives include: (1) designing reliable...
- This $211,129 Project Grant was awarded on January 1, 2025 by the National Science Foundation (NSF) under the Computer and Information Science and Engineering (CFDA 47.070) program to Northeastern University. The project, titled "CAREER: NEURAL NETWORK-INSPIRED INFORMATION PROCESSING BEYOND THE BINARY DIGITAL ABSTRACTION", aims to develop a design framework for efficient information processing using non-binary representations and in-memory memory/computation. Key research thrusts...
- This three-year National Science Foundation Project Grant of $418,907 will support the development of an energy-efficient, customizable mixed-signal deep neural network system using a novel reconfigurable crossbar architecture and new memory technology at Wayne State University. Funded through the NSF's Engineering program (CFDA 47.041), the university aims to address challenges in reliably implementing analog computation for deep learning through innovations at the technology, circuit,...
- The University of Michigan will receive $393,983 from the National Science Foundation under a three-year Project Grant to develop scaled non-volatile bulk analogue memory for neuromorphic computing. The NSF Division of Electrical, Communications and Cyber Systems awarded the grant through its Engineering program (CFDA 47.041), which seeks to foster innovation in engineering research. Specifically, the University will work to advance memory technologies that can support neuromorphic computing...
- This $500,000 National Science Foundation project grant supports research at the University of California, Riverside to develop novel machine learning-based electromigration analysis and optimization methods for very large-scale integrated circuit design. Specifically, the university will explore enhanced physics-informed neural network approaches for multi-segment interconnect stress analysis and full-chip electromigration-induced voltage drop modeling. Researchers will also develop efficient...
- This National Science Foundation project grant of $600,000 supports research through June 2025 at the University of Notre Dame under the Office of International Science and Engineering program (CFDA 47.079). The grant funds a partnership between researchers at Notre Dame and Ecole Centrale de Lyon in France to study the impact of emerging information processing technologies on computer architectures and applications. Specifically, the researchers will examine how new architectures enabled by...
- This $1,139,000 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) aims to develop innovative dataflow processors that can enhance the energy efficiency and security of edge computing systems. The key technical innovations include a parallel dataflow representation of programs and a simplified, scalable spatial architecture for dataflow processors. The research will be conducted by a diverse team at...
This $193,176 three-year Project Grant from the National Science Foundation's Division of Electrical, Communications and Cyber Systems, under the Engineering (47.041) federal grant program, will fund research at Villanova University to develop approximate computing techniques. The university researchers will create input-aware error models for approximate circuits considering data impacts. They will also develop a graph neural network framework to estimate application quality and a resource-aware configuration tool to optimize performance and energy while meeting quality constraints. The tools and models developed over the course of this research are intended to assess approximate computing approaches and automatically configure them, with the goal of enabling wider adoption of these energy-efficient computing methods.
Mod # | Description | ReasonForModification | Federal Obligation | Date |
|---|---|---|---|---|
| Not listed | $193.2k | 8/23/22 |