EurekAlert! News by Subject: Mathematics - News by Subject: Mathematics

Mathematics

Updates every hour. Last Updated: 11-May-2025 06:09 ET (11-May-2025 10:09 GMT/UTC)

Treatment expectancies and psilocybin vs escitalopram for depression

JAMA Network
Peer-Reviewed Publication

About The Study: This randomized controlled trial secondary analysis examines the association between treatment expectancies and the relative efficacy of psilocybin compared with escitalopram for major depressive disorder.

Journal: JAMA Psychiatry

Department of Energy announces $36 million for student traineeships

DOE/US Department of Energy
Grant and Award Announcement

Today, the U.S. Department of Energy (DOE) announced 29 projects totaling $36 million to 42 institutions in 16 states for traineeships for undergraduate students, graduate students, and postdoctoral researchers in science, technology, engineering, and mathematics (STEM). The funding, through the DOE Office of Science’s RENEW initiative, will support hands-on research experience, professional development activities to build or reinforce STEM identity, and mentorship to support personal and professional growth of trainees.

Advances in systemic mastocytosis: Early signs of efficacy for a novel drug and a new predictive tool to guide patient care

Dana-Farber Cancer Institute
Reports and Proceedings

Dana-Farber Cancer Institute researchers are breaking new ground in the fight against systemic mastocytosis, a rare and heterogeneous condition that is often challenging for clinicians to accurately diagnose and treat.

Space-time crystals, an important step toward new optical materials

Karlsruher Institut für Technologie (KIT)
Peer-Reviewed Publication

Photonic space-time crystals are materials that could increase the performance and efficiency of wireless communication or laser technologies. They feature a periodic arrangement of special materials in three dimensions as well as in time, which enables precise control of the properties of light. Working with partners from Aalto University, the University of Eastern Finland and Harbin Engineering University in China, scientists from the Karlsruhe Institute of Technology (KIT) have shown how such four-dimensional materials can be used in practical applications. They published their results in Nature Photonics (DOI: 10.1038/s41566-024-01563-3).

Journal: Nature Photonics

Black-box forgetting: A new method for tailoring large AI models

Tokyo University of Science
Reports and Proceedings

Pretrained large-scale AI models need to ‘forget’ specific information for privacy and computational efficiency, but no methods exist for doing so in black-box vision-language models, where internal details are inaccessible. Now, researchers from Japan addressed this issue through an innovative strategy based on latent context sharing, successfully getting an image classifier to forget multiple classes it was trained on. Their findings could expand the use cases of large-scale AI models while safeguarding end users’ privacy.

Funder: NEC Corporation
Meeting: NeurIPS 2024

Not so simple machines: Cracking the code for materials that can learn

University of Michigan
Peer-Reviewed Publication

University of Michigan physicists have devised an algorithm that provides a mathematical framework for how learning works in lattices called mechanical neural networks.

Journal: Nature Communications
Funder: U.S. National Science Foundation, Office of Naval Research

New AI cracks complex engineering problems faster than supercomputers

Johns Hopkins University
Peer-Reviewed Publication

Modeling how cars deform in a crash, how spacecraft responds to extreme environments, or how bridges resist stress could be made thousands of times faster thanks to new artificial intelligence that enables personal computers to solve massive math problems that generally require supercomputers.

Journal: Nature Computational Science

81
82
83
84
85