Zhou receives funding for novel performance profiling & analysis infrastructure for scientific deep learning workloads
2025-08-05
(Press-News.org)
Zhou Receives Funding For Novel Performance Profiling & Analysis Infrastructure For Scientific Deep Learning Workloads
Keren Zhou, Assistant Professor, Computer Science, College of Engineering and Computing (CEC), received funding for the project: “Collaborative Research: Elements: DLToolkit: A Novel Performance Profiling and Analysis Infrastructure for Scientific Deep Learning Workloads.”
The rapid adoption of deep learning (DL)-driven artificial intelligence (AI) applications makes it more crucial than ever to fully exploit available computational power to accelerate AI innovation. He is developing innovative performance measurement and analysis techniques tailored for scientific DL workloads and encapsulating these techniques into a profiling infrastructure (i.e., DLToolkit).
Building upon existing open-source performance tools, Zhou holds the DLToolkit will offer scalable analysis, aggregation, and visualization of scientific DL workloads, and provide invaluable insights to domain scientists. Zhou holds the toolkit will thus foster expedited innovation in scientific applications using DL.
Zhou received $274,265 from the National Science Foundation for this project. Funding began in June 2025 and will end in late May 2028.
###
ABOUT GEORGE MASON UNIVERSITY
George Mason University is Virginia’s largest public research university. Located near Washington, D.C., Mason enrolls more than 40,000 students from 130 countries and all 50 states. Mason has grown rapidly over the past half-century and is recognized for its innovation and entrepreneurship, remarkable diversity, and commitment to accessibility. In 2023, the university launched Mason Now: Power the Possible, a one-billion-dollar comprehensive campaign to support student success, research, innovation, community, and stewardship. Learn more at gmu.edu.
END
ELSE PRESS RELEASES FROM THIS DATE:
2025-08-05
Sleeter Receives Funding For Revolutionary War Teaching Guides
Nathan Sleeter, Research Assistant Professor, History and Art History, Roy Rosenzweig Center for History and New Media (RRCHNM), College of Humanities and Social Sciences (CHSS), received funding for: “Revolutionary War Teaching Guides.”
Sleeter will develop the teaching guides which use Library of Congress primary sources to support educators on the topic of Black and Indigenous Americans during the Revolutionary War. Sleeter will also engage in outreach to promote the guides with educators.
He will use sources such as diaries, ...
2025-08-05
The research and design of an adjustable multispectral compatible infrared camouflage device based on the infrared radiation characteristics of Rosaceae plants, aims to achieve multifunctional compatibility of infrared camouflage, thermal management, laser stealth, and visible light camouflage. The device design employs a particle swarm optimization algorithm combined with the finite difference time domain method to obtain structural parameters: Cr/In3SbTe2(IST)/Ge/TiO2/Ge/ZnS (top cylindrical).Through experimental verification, it is demonstrated that in the amorphous state (aIST), the device can achieve simulated plant ...
2025-08-05
Professor Xufeng Jing’s research team at China Jiliang University has conducted a systematic study on wireless communication technologies based on metasurfaces. This paper provides a detailed introduction to the working principles and classifications of passive, active, and semi-active metasurfaces, with a particular focus on how digital-coded metasurfaces achieve precise control over the phase and polarization of electromagnetic waves through dynamic tuning of unit structures.
The research team emphasizes the core advantages of metasurfaces in wireless communication, including miniaturization, ...
2025-08-05
Greenacres, FL and London, UK — July 31, 2025
Tharos Ltd today announced encouraging exploratory findings from a small, uncontrolled pilot evaluation of its enzyme‑rich malt extract (ERME), marketed as EquiNectar® for horses and CaniNectar® for dogs, and sold since 2018 as an animal feed supplement. Over four weeks, stool 16S rRNA gene sequencing showed a directional decline in low‑abundance sequence reads annotated to the genus Borrelia in a subset of animals. While stool‑based, genus‑level annotations are not diagnostic for Lyme disease or a measure of systemic organism burden, the coherence of these signals—together ...
2025-08-05
Self-driving cars know their own way in unpredictable traffic, thanks to path planning technology. Among current AI-driven efforts to make path planning more efficient and reliable, a research team has developed an optimization method proven especially effective in uncertain environments. The results were published June 3 under the title “Action-Curiosity-Based Deep Reinforcement Learning Algorithm for Path Planning in a Nondeterministic Environment” in Intelligent Computing, a Science Partner Journal.
The team evaluated their method in a realistic ...
2025-08-05
As use of the popular anti-diabetic and weight-loss drug Ozempic skyrockets, so have concerns about the medication’s side effects. One such side effect is loss of “lean mass”—body weight that isn’t fat—raising concerns that Ozempic could be reducing muscle mass and strength.
New research in mice suggests that muscle mass changes less than expected, but muscles may still get weaker, pointing out an urgent need for clinical studies to pin down the full effects of the popular medications.
“If we want to really help the individuals ...
2025-08-05
About The Study: The findings of this study suggest that policies for insurance coverage of supplemental screening based on breast density may have limited ability to improve early detection for Black women.
Corresponding Author: To contact the corresponding author, Anne Marie McCarthy, ScM, PhD, email annemcc@pennmedicine.upenn.edu.
(doi:10.1001/jamanetworkopen.2025.25216)
Editor’s Note: Please see the article for additional information, including other authors, author contributions and affiliations, ...
2025-08-05
About The Study: In this cross-sectional study, structural disparities in rapid emergency medical services (EMS) access were associated with historically redlined areas. Strategic resource allocation and system redesign are warranted to address these inequities in prehospital emergency care.
Corresponding Author: To contact the corresponding author, Cherisse Berry, MD, email cherisse.berry@rutgers.edu.
To access the embargoed study: Visit our For The Media website at this link https://media.jamanetwork.com/
(doi:10.1001/jamanetworkopen.2025.25681)
Editor’s ...
2025-08-05
A critical water source for vast areas of the Northern Territory is drying at an accelerating rate, according to new findings, with losses clearly visible from space.
The Cambrian Limestone Aquifer (CLA) is a large, interconnected limestone system containing high-quality groundwater that supports numerous NT rivers, towns, Indigenous communities, pastoral enterprises, and irrigated agriculture.
But the findings, published in a new study led by Griffith University researchers, show the aquifer has experienced significant water loss since 2014, reaching its lowest recorded storage level in 2021 (the end of the study period).
The study draws on two decades ...
2025-08-05
What looks like a microscopic dance battle is actually a life-or-death strategy. In scalding hot water rushing through narrow channels, some bacteria have evolved a surprising survival technique: they cling to surfaces, stand upright, and sway rhythmically—like tiny street dancers fighting the flow. Watch the video of the bacterial “reverse-flow dance”: https://youtu.be/JDN28g-aE78
This dramatic behavior was captured for the first time on video by a team led by Dr. Daisuke Nakane, Associate Professor at the University of Electro-Communications in Tokyo, in collaboration with Professors Masatada Tamakoshi and Ryota Morikawa at ...
LAST 30 PRESS RELEASES:
[Press-News.org] Zhou receives funding for novel performance profiling & analysis infrastructure for scientific deep learning workloads
