Articles tagged with Algorithms
A team of researchers developed a new technique combining methods to simulate molecules, achieving accuracy and efficiency on the Frontier exascale supercomputer. They broke records with simulations of over one million electrons and scaled their algorithm to an EFlop/s processing quintillion calculations per second.
Researchers developed a machine-learning tool that provides accurate predictions for flood-prone areas, using historical data and weather-based predictors. The model can predict short-term river discharge with high accuracy, giving real-time data on water movement through the river.
A new approach using topological data analysis (TDA) enhances the reliability and reduces bias in AI systems used for medical diagnosis in radiology. TDA captures intricate features and provides a holistic view of medical images, leading to more accurate diagnoses and equitable patient care.
A new study reveals that an AI algorithm can accurately detect early-stage metabolic-associated steatotic liver disease (MASLD) in patients who meet the criteria, leaving 83% undiagnosed. This highlights the need for improved screening and diagnosis methods to prevent progression to advanced liver disease.
Researchers at Graz University of Technology have developed a digital monitoring system to prevent costly mistakes in concreting processes. The system uses sensors and algorithms to measure and analyze various parameters, providing real-time warnings for potential issues, and eliminating the need for rework.
A WVU research partnership with the DEA aims to improve fast and accurate identification of psychoactive substances like fentanyl. The Expert Algorithm for Substance Identification (EASI) will enable labs using different instruments to share data on chemical profiles, helping to identify drugs like fentanyl.
HemaChrome's machine learning-based technology enables instant and noninvasive measurement of blood hemoglobin levels from digital photos, facilitating point-of-care diagnostic tests. The collaboration with Global Health Labs aims to address anemia diagnosis gaps in low- and middle-income countries.
The study emphasizes the need for a coherent approach to understanding A.I. threats, recognizing the intricate interplay between technology and society. Experts propose involving laypeople and experts in risk assessment processes, as well as promoting social resilience to ensure better decision-making.
Researchers at Lehigh University are using advanced algorithms and cross-domain data to help cities predict human movement patterns, enabling better planning and preparedness for events and emergencies. The model will account for variations in data streams from different sources, such as cell towers, GPS, and financial transactions.
Researchers found that X/Twitter's algorithm presents users with milder and less polarizing information than chronological timeline news. However, this has implications for the use of these platforms to find trustworthy news, as users question its credibility even when coming from legitimate sources.
Professor Ruth Britto and her international team will develop new algorithmic methods with applications in mathematics, particle physics, and gravity. They aim to tackle longstanding computational bottlenecks and push the boundaries of numerous areas of theoretical physics.
Researchers have summarized advanced field weakening (FW) control strategies for permanent magnet synchronous motors (PMSMs), including offline calculation methods, online computational methods, and model predictive control (MPC)-related methods. The studies highlight the importance of balancing computational difficulty with control ro...
A team of MIT engineers developed an algorithm to identify causal links in complex systems, taking data from various sources and analyzing interactions between variables. The method generates a causality map linking variables with likely cause-and-effect relationships, including synergistic and redundant links.
The team developed an exascale climate emulator with enhanced resolution without increasing data storage needs. The emulator offers a remarkable resolution of 3.5 kilometers, replicating local conditions on a timescale from days to hours.
A Cornell University research team found that strategically placing a mix of medium-speed and fast-charging stations in urban areas increases driver usage and improves investor profitability by 50-100%. The team used Bayesian optimization to analyze data from Atlanta, taking into account factors like traffic and road characteristics.
A new training algorithm called ternarized gradient BNN (TGBNN) enables learning capabilities for binarized neural networks (BNNs) on IoT edge devices. The proposed MRAM-based CiM architecture achieves faster convergence and matching accuracy with regular BNNs.
Researchers developed a machine learning model to predict dielectric function of materials, facilitating novel dielectric material development. The model speeds up calculations by using chemical bonds between atoms and achieving accuracy close to first-principle calculations.
The GALAD algorithm combines AFP, AFP-L3, and PIVKA-II for HCC detection. Combining GAAD, GALAD, or PIVKA-II with ultrasound improves diagnostic efficiency compared to recommended strategies.
Researchers have developed Rastermap, a visualization tool that enables scientists to uncover activity patterns in thousands of neurons. The tool sorts neuronal activity into clusters based on similarity and maps them onto a graphical representation, allowing for the identification of patterns that can be further tested in the lab.
A UVA professor has developed a new computational algorithm to find tightly connected clusters, or triangle-dense subgraphs, within large networks. This breakthrough can help uncover suspicious activity in fraud detection and identify community dynamics on social media with greater precision.
Researchers at the University of Copenhagen's Quantum for Life Centre have developed a new mathematical recipe to make quantum simulators more scalable and efficient. This breakthrough could speed up the development of new medicines from years to months by predicting how molecules behave in the human body before laboratory trials.
A new method called Clio allows robots to make task-relevant decisions by identifying the parts of a scene that matter. In real experiments, Clio successfully mapped scenes at different levels of granularity based on natural-language prompts and enabled robots to grasp objects of interest.
A new vehicle allocation strategy for dockless bike-sharing systems at night can improve the balance of urban transportation and reduce traffic congestion. The approach uses a three-way classification method and behavioral decision theory to forecast demand and optimize resource allocation, reducing waste and enhancing system efficiency.
A team of researchers from UMass Amherst debunks the idea that Facebook's algorithms successfully filtered out untrustworthy news during the 2020 election. The study found that temporary changes to the algorithm were not accounted for, leading to misperceptions about the platform's reliability.
Researchers found that large language models used in home surveillance can make inconsistent decisions about calling the police, even when videos show no crime. Models often disagreed with each other and exhibited inherent biases influenced by neighborhood demographics.
Researchers at Klick Labs developed an AI technique using vocal biomarkers to predict chronic high blood pressure with up to 84% accuracy. The study used machine learning to analyze hundreds of indiscernible vocal biomarkers, including pitch variability and speech energy distribution patterns.
A breakthrough technology allows for touchless infrared imaging to monitor changes in pupil size and gaze direction behind closed eyes. This innovation can help identify wakefulness, awareness, and pain in sleep, anesthesia, and intensive care, enabling more accurate clinical decision-making.
Researchers Maria Eichlseder and Fariba Karimi will study keyless encryption and AI's impact on online social networks to promote fair algorithms. Their projects aim to address open problems in cryptographic systems and quantify intersectional inequality.
A new algorithm, inspired by the nervous system's matchmaker, pairs drivers with riders in a way that maximizes everyone's happiness. The algorithm creates near-optimal pairings while preserving privacy, making it suitable for everyday applications.
Researchers questioned the Cascadia subduction zone's earthquake record, finding that turbidite layers showed no better correlation than random chance. The study suggests a need for further research on turbidite layers and their connection to past earthquakes.
Researchers at Boston University created an AI tool that can determine the cause of dementia using commonly collected patient data, boosting doctor accuracy by 26%. The algorithm identifies 10 types of dementia, including vascular and frontotemporal dementia, to help doctors manage patients more effectively.
A recent study found that approximately half of FDA-approved AI medical devices are not trained on real patient data, sparking concerns about device accuracy. The researchers analyzed 500+ medical AI devices and discovered that many lacked clinical validation data, which is essential for ensuring the credibility of these technologies.
The researchers will develop new algorithms to identify clusters within large datasets, enabling better community detection. They plan to test the method in various applications, including single-cell genomics and scientometrics.
A new AI-based digital platform has been developed to analyze tissue sections from lung cancer patients, making diagnosis faster and more accurate. The platform uses algorithms that enable fully automated analysis of digitized tissue samples, allowing for personalized therapy based on molecularly specific genetic changes.
Researchers at Washington State University developed an AI algorithm that optimizes 3D printing settings, reducing time and cost for engineers. The algorithm improved the accuracy and quality of printed models, particularly for complex biomedical devices like kidneys and prostates.
A new algorithm developed at Washington State University improves safety and efficiency in robots working with humans by accounting for human carelessness. The tool has shown a maximum improvement of 80% in safety and 38% in efficiency compared to existing methods, and the researchers plan to test it in real-world settings.
A computer algorithm has achieved a 98% accuracy in predicting different diseases by analyzing the color of the human tongue. The proposed imaging system can diagnose various health conditions, including diabetes, stroke, and COVID-19, using a simple and affordable method.
Researchers develop an unsupervised deep learning-based method to reconstruct particle distribution in Tomographic PIV, achieving superior performance over traditional methods. The new technique demonstrates potential for practical applications in high-density particle fields and high-velocity flow fields.
Researchers have discovered a non-memory-based mechanism for animals to cache and retrieve food, challenging long-held beliefs about animal cognition. The proposed mechanism uses neural networks similar to hash functions, allowing for efficient storage and retrieval of cache locations.
A new study led by CU Boulder computer scientist Theodora Chaspari found that AI algorithms can be confused by natural variations in speech patterns between different genders and races. This can lead to underdiagnosis or misdiagnosis of mental health concerns like depression.
Researchers at Pohang University of Science & Technology have developed a novel analog hardware using ECRAM devices that maximizes AI computational performance. Their technique, which uses a three-terminal structure with separate paths for reading and writing data, demonstrates excellent electrical and switching characteristics.
Researchers have introduced a new AI calibration method called Thermometer, which enables efficient calibration of large language models for various tasks. This technique leverages temperature scaling to adjust a model's confidence and can generalize to new tasks without requiring additional labeled data.
The University of Leicester is developing a method to shrink artificial intelligence algorithms for smarter spacecraft. The REALM project aims to demonstrate streamlined machine learning algorithms suitable for limited spacecraft power and computing performance.
A machine learning algorithm was trained to predict individuals with functional neurological disorder (FND) by analyzing their brain structure. The algorithm achieved significant above-chance accuracy in classifying FND participants against healthy controls and psychiatric samples, highlighting the importance of considering both brain ...
Researchers found that large language models perform poorly in high-stakes situations despite being better than smaller models, due to misalignment with human generalization function. Human generalization, which involves forming beliefs about others' abilities, plays a significant role in LLM performance and deployment.
A new position paper argues that single race-agnostic FRAX models would unfairly discriminate against Black, Asian, and Hispanic communities. The authors recommend retention of ethnic and race-specific FRAX models for the US with updated data on fracture and death hazards.
Researchers at USC developed a new method to accurately predict wildfire spread using satellite data and artificial intelligence. The model offers a potential breakthrough in wildfire management and emergency response, providing more precise and timely data for firefighters and evacuation teams battling wildfires.
A study found that large language models, despite accuracy in medical exams, fail to consistently request necessary examinations and often deviate from treatment guidelines. In comparison to human doctors, AI diagnoses achieved lower accuracy rates, highlighting concerns about their suitability for everyday clinical practice.
A new study reveals significant discrepancies in the effectiveness of political ads on Facebook and Instagram, favoring more extremist groups. Over 70% of parties used user profiling in their ads, and the far-right AfD proved to be the most effective, with ads almost six times more efficient than competitors.
A team of UCSF specialists predicts 24-hour seizure risk using brain activity patterns that foreshadow seizures. The discovery may improve quality of life for 2.9 million Americans living with epilepsy.
Researchers developed a method to assess the reliability of foundation models, enabling users to choose the best model for their task without testing it on real-world data. The approach measures consensus among multiple models and aligns representations to compare consistency.
Scientists from Trinity College Dublin created a computer program that visualizes molecular structure in the style of Piet Mondrian. The program uses blocks of color to represent symmetry and shape, making it easier to understand complex molecular interactions.
GenSQL integrates a tabular dataset and a generative probabilistic AI model to analyze complex tabular data. It can detect anomalies, predict outcomes, and generate synthetic data with just a few keystrokes.
Researchers developed an AI model that can estimate lung function from chest radiographs with high accuracy, potentially expanding options for pulmonary function assessment in patients who have difficulty performing spirometry. The study found a remarkably high agreement rate between the AI model's estimates and actual spirometric data.
A new machine learning-based method uses 3D structure of protein backbone with large language models to predict molecular changes that lead to better antibody drugs. The approach resulted in a 25-fold improvement against a virus, outperforming traditional methods that rely on generating huge amounts of data about protein sequences.
The SDC-DeepLabv3+ algorithm achieved a mean pixel accuracy of 95.84% and mean intersection over union of 96.87%, reducing background interference and enhancing filament visibility. The method shows potential for improved harvesting robot performance and precise filament harvesting.
A study found that large language models (LLMs) like ChatGPT underperform state-of-the-art detectors but can explain their analysis in plain language. LLMs' semantic knowledge makes them well-suited for detecting deepfakes, providing a common sense understanding of reality.
Researchers developed an AI model that accurately predicts metal yield strength by combining physical theory with machine learning. The model outperforms traditional methods, which often rely on extensive experimentation.
Researchers at NTNU have developed a way to reconstruct 3D models of the colon using single images taken by capsule endoscopy cameras. These models can aid in detecting abnormalities and signs of disease, enabling specialists to make diagnoses faster.
A new study found that animals use a wide range of strategies to accomplish tasks, many of which are just as effective as the optimal solution but require less brain power. The research provides a theoretical framework for understanding these 'good enough' strategies and their potential applications in animal behavior.