Showing 20 articles starting at article 181
< Previous 20 articles Next 20 articles >
Categories: Mathematics: Modeling
Published Self-supervised AI learns physics to reconstruct microscopic images from holograms
(via sciencedaily.com) Researchers have unveiled an artificial intelligence-based model for computational imaging and microscopy without training with experimental objects or real data. The team introduced a self-supervised AI model nicknamed GedankenNet that learns from physics laws and thought experiments. Informed only by the laws of physics that universally govern the propagation of electromagnetic waves in space, the researchers taught their AI model to reconstruct microscopic images using only random artificial holograms -- synthesized solely from 'imagination' without relying on any real-world experiments, actual sample resemblances or real data.
Published How good is that AI-penned radiology report?
(via sciencedaily.com) New study identifies concerning gaps between how human radiologists score the accuracy of AI-generated radiology reports and how automated systems score them. Researchers designed two novel scoring systems that outperform current automated systems that evaluate the accuracy of AI narrative reports. Reliable scoring systems that accurately gauge the performance of AI models are critical for ensuring that AI continues to improve and that clinicians can trust them.
Published Deep learning for new protein design
(via sciencedaily.com) Deep learning methods have been used to augment existing energy-based physical models in 'do novo' or from-scratch computational protein design, resulting in a 10-fold increase in success rates verified in the lab for binding a designed protein with its target protein. The results will help scientists design better drugs against diseases like cancer and COVID-19.
Published New method simplifies the construction process for complex materials
(via sciencedaily.com) A new technique incorporates many different building blocks of cellular metamaterials into one unified graph-based representation. This can be used to make a user-friendly interface that can quickly and easily model metamaterials, edit the structures, and simulate their properties.
Published Scientists uncover a surprising connection between number theory and evolutionary genetics
(via sciencedaily.com) An interdisciplinary team of mathematicians, engineers, physicists, and medical scientists has uncovered an unexpected link between pure mathematics and genetics, that reveals key insights into the structure of neutral mutations and the evolution of organisms.
Published Researchers successfully train a machine learning model in outer space for the first time
(via sciencedaily.com) Original source Scientists have trained a machine learning model in outer space, on board a satellite. This achievement could revolutionize the capabilities of remote-sensing satellites by enabling real-time monitoring and decision making for a range of applications.
Published A simpler method for learning to control a robot
(via sciencedaily.com) A new machine-learning technique can efficiently learn to control a robot, leading to better performance with fewer data.
Published AI predicts the work rate of enzymes
(via sciencedaily.com) Enzymes play a key role in cellular metabolic processes. To enable the quantitative assessment of these processes, researchers need to know the so-called 'turnover number' (for short: kcat) of the enzymes. A team of bioinformaticians now describes a tool for predicting this parameter for various enzymes using AI methods.
Published The economic life of cells
(via sciencedaily.com) A team has combined economic theory with biology to understand how natural systems respond to change. The researchers noticed a similarity between consumers' shopping behavior and the behavior of metabolic systems, which convert food into energy in our bodies. The team focused on predicting how different metabolic systems might respond to environmental change by using an economic tool called the Slutsky equation. Their calculations indicated that very different metabolic systems actually share previously unknown universal properties, and can be understood using tools from other academic fields. Metabolic processes are used in drug development, bioengineering, food production and other industries, so being able to predict how such systems will respond to change can offer many benefits.
Published Supercomputer used to simulate winds that cause clear air turbulence
(via sciencedaily.com) Using Japan's most powerful supercomputer, researchers reproduced cases of clear air turbulence around Tokyo. They simulated the fine vortices responsible for this dangerous phenomenon. The usefulness of the simulation in predicting turbulence was confirmed by comparing simulation data with data from aircraft recordings. This research should improve the forecasting of turbulence.
Published Researcher turns one of the basic rules of construction upside down
(via sciencedaily.com) Original source Structural engineers are familiar with seventeenth-century scientist Robert Hooke's theory that a hanging chain will mirror the shape of an upstanding rigid arch. However, new research now shows that this common-held belief is incorrect because, regardless of the similarities, the hanging chain and the arch are two incompatible mechanical systems.
Published Board games are boosting math ability in young children
(via sciencedaily.com) Board games based on numbers, like Monopoly, Othello and Chutes and Ladders, make young children better at math, according to a comprehensive review of research published on the topic over the last 23 years.
Published Machine learning takes materials modeling into new era
(via sciencedaily.com) The arrangement of electrons in matter, known as the electronic structure, plays a crucial role in fundamental but also applied research such as drug design and energy storage. However, the lack of a simulation technique that offers both high fidelity and scalability across different time and length scales has long been a roadblock for the progress of these technologies. Researchers have now pioneered a machine learning-based simulation method that supersedes traditional electronic structure simulation techniques. Their Materials Learning Algorithms (MALA) software stack enables access to previously unattainable length scales.
Published Deciphering the thermodynamic arrow of time in large-scale complex networks
(via sciencedaily.com) A solution for temporal asymmetry -- or entropy production -- in thermodynamics has been developed to further our understanding of the behavior of biological systems, machine learning, and AI tools. The researchers worked on the time-irreversible Ising model dynamics caused by asymmetric connections between neurons.
Published Researchers teach an AI to write better chart captions
(via sciencedaily.com) Original source A new tool helps scientists develop machine-learning models that generate richer, more detailed captions for charts, and vary the level of complexity of a caption based on the needs of users. This could help improve accessibility for people with visual disabilities.
Published Evaluating cybersecurity methods
(via sciencedaily.com) Original source Researchers created a generic framework that enables an engineer or scientist to evaluate the effectiveness of defense schemes that seek to limit a hacker's ability to learn secret information by observing the behavior of a victim computer program.
Published Turning old maps into 3D digital models of lost neighborhoods
(via sciencedaily.com) Original source Imagine strapping on a virtual reality headset and 'walking' through a long-gone neighborhood in your city -- seeing the streets and buildings as they appeared decades ago. That's a very real possibility now that researchers have developed a method to create 3D digital models of historic neighborhoods using machine learning and historic Sanborn Fire Insurance maps.
Published Generative AI models are encoding biases and negative stereotypes in their users
(via sciencedaily.com) Original source In the space of a few months generative AI models, such as ChatGPT, Google's Bard and Midjourney, have been adopted by more and more people in a variety of professional and personal ways. But growing research is underlining that they are encoding biases and negative stereotypes in their users, as well as mass generating and spreading seemingly accurate but nonsensical information. Worryingly, marginalized groups are disproportionately affected by the fabrication of this nonsensical information.
Published Bridging traditional economics and econophysics
(via sciencedaily.com) Original source How do asset markets work? Which stocks behave similarly? Economists, physicists, and mathematicians work intensively to draw a picture but need to learn what is happening outside their discipline. A new paper now builds a bridge.
Published High-quality child care contributes to later success in science, math
(via sciencedaily.com) Original source Children who receive high-quality child care as babies, toddlers and preschoolers do better in science, technology, engineering and math through high school, and that link is stronger among children from low-income backgrounds, according to new research.