Articles tagged with Algorithms
Computer scientists designed a reconstruction attack that proves US Census data can be exposed and stolen with current privacy measures. The study demonstrates risks to individual respondents' privacy, highlighting the need for differential privacy techniques to protect sensitive information.
A new algorithm uses patient-reported outcomes to identify patients with cancer who are at increased odds of having unplanned emergency department visits. The algorithm, which calculates a symptom complexity score, has shown that patients with high complexity scores are three times more likely to use the ED.
A recent study published in Human Reproduction found a significant association between prenatal alcohol exposure and changes in children's facial features. The research used AI to analyze three-dimensional images of children's faces at ages nine and thirteen, revealing that mothers' alcohol consumption before and during pregnancy can h...
Researchers at MIT developed a technique to improve machine-learning models' reliability without requiring additional data or extensive computing resources. The method uses a simpler companion model to estimate uncertainty, enabling more effective uncertainty quantification.
Researchers at Brigham Young University have developed an AI algorithm that can accurately locate players and determine formations in football game footage. The system has achieved over 90% accuracy on player detection and 85% accuracy on formation identification, potentially eliminating the need for manual annotation and analysis.
Researchers tested three common techniques to make algorithms fairer and found that one approach didn't reduce social norm bias at all. They proposed a new technique: a formula to directly measure social norm bias in an algorithm so it can be corrected. This bias can persist even after overt discrimination is removed.
Researchers at Osaka Metropolitan University developed a new method to evaluate X-ray microbeam diameter using mathematical analysis, outperforming conventional methods. The uniform evaluation method is expected to be widely adopted as an international standard.
Researchers utilized the Chemistry42 platform to generate novel molecular structures and identified a hit molecule for CDK20, a promising target for hepatocellular carcinoma. The platform's customizable reward function and generative models enabled efficient design and optimization of molecules.
Researchers developed an ethical algorithm to distribute risk fairly in autonomous vehicles, considering multiple factors such as responsibility and traffic regulations. The new approach enables more nuanced decision-making in critical situations, weighing risks and avoiding aggressive maneuvers.
Researchers at Bar-Ilan University have developed a new type of artificial neural network that outperforms traditional deep learning architectures. By using tree architecture with single routes to output units, they achieve better classification success rates, paving the way for efficient and biologically-inspired AI hardware.
Scientists at Göttingen University have created a novel approach to produce X-ray images in color from a single exposure, eliminating the need for focusing and scanning. This breakthrough method uses an X-ray color camera and a specially structured plate to capture the intensity pattern of fluorescing atoms in a sample.
Lithuanian researchers have developed an AI system to detect early symptoms of Parkinson's disease using voice data. The study found that patients with early-stage Parkinson's often speak in a quieter, more monotonous tone, which can be difficult to notice by ear.
A new Swedish quantum computer is being made available to the industry, accompanied by a test bed and a quantum helpdesk. The test bed will allow companies and researchers to solve problems using quantum technology at a significantly lower cost than existing commercial options.
University of Illinois scientists developed a way to accurately map tilled land in real-time using ground, airborne, and satellite imagery. The method increased mapping accuracy by 67% compared to upscaling directly from the ground to the satellite.
Researchers use coherent correlation imaging to image the evolution of magnetic domains in time and space without prior knowledge. The study reveals thermal motion and pinning effects on domain boundaries, unlocking new insights into magnetism's microcosm.
Researchers use drone technology to detect changes in building facades and assess technical condition, enabling accurate classification and identification. The system can also be used for urban heritage site monitoring and planning urban development.
Researchers at Brookhaven National Laboratory have successfully discovered new materials using artificial intelligence and self-assembly. The AI-driven technique led to the discovery of three new nanostructures, expanding the scope of self-assembly's applications in microelectronics and catalysis.
Researchers developed machine learning models to accurately calculate fine particulate matter in urban air pollution using AI and traffic data. The models provide a high-resolution estimation of city street pollution surface, enabling transportation and epidemiology studies to assess health impacts.
Scientists have developed AI techniques to track crevasses on the Thwaites Glacier, which could impact global sea levels by up to 60cm. The study found a complex interplay between crevasse formation and ice flow speed changes.
A team of researchers from Xi'an Jiaotong-Liverpool University and other institutions has identified a flexible and user-friendly model for predicting flood frequency in a changing environment. The fractional polynomial-based regression method is more effective than existing models, which often fail to account for factors like climate ...
Researchers used machine learning to analyze parole data from New York and found that the New York State Parole Board could safely grant parole to more inmates, potentially doubling the release rate. The study suggests eliminating racial disparities in release rates while maintaining public safety.
Researchers developed a new method using Bayesian estimation to reduce X-ray fluorescence analysis time from 7 seconds to 3 seconds per measurement point, saving up to 11 hours for large-scale elemental distributions.
Researchers at Drexel University used GPT-3 to spot early signs of Alzheimer's in spontaneous speech, achieving 80% accuracy. The program analyzed word-use, sentence structure and meaning from transcripts to identify characteristic profiles of Alzheimer's speech.
A new algorithm developed by researchers at Charité improves motor symptoms comparable to standard of care treatment, increasing efficiency in deep brain stimulation for Parkinson's disease. The study suggests a promising result for imaging-based algorithms to simplify clinical practice and improve therapeutic outcomes.
A recent study has identified common and unique cellular processes in six neurodegenerative diseases, providing new insights into the underlying causes of these conditions. The research used machine learning analysis to compare RNA markers in whole blood samples from patients with distinct diseases, revealing eight shared themes across...
New signal-processing algorithms have been shown to help mitigate the impact of turbulence in free-space optical experiments. The researchers achieved record results using commercially available photonic lanterns and a spatial light modulator to emulate turbulence.
Researchers from Brown and MIT developed a new framework that uses machine learning and sequential sampling to predict rare disasters like earthquakes and pandemics with less data. The framework, called DeepOnet, has been shown to outperform traditional modeling efforts in predicting scenarios, probabilities and timelines of rare events.
A team of researchers from Tokyo University of Science developed a super-hierarchical and explanatory analysis method for magnetic reversal processes, enabling the detection of subtle microscopic changes. The new algorithm can predict stable/metastable states in advance and improve the reliability of spintronics devices.
A University of Houston researcher has developed a method to describe complex systems using the least number of variables possible, reducing complexity from millions to just one. This advancement speeds up science with efficiency and ability to understand and predict natural system behavior.
A new, non-invasive malaria detection tool developed by a University of Queensland-led team can quickly identify entire villages or towns suffering from the disease. The device uses infrared-light and is chemical-free, needle-free, and detects malaria through the skin.
Researchers develop an algorithm to automate feature extraction from HD maps and point cloud data. The model achieved high precision in detecting road signs and traffic lights, with zero false detections.
Researchers at MIT have developed a scheme for private information retrieval that is about 30 times faster than other comparable methods. The technique enables users to search an online database without revealing their query to the server, with potential applications in private communication and targeted advertising.
A team of researchers from the University of Pennsylvania has developed a new algorithm, metadynamics, that can navigate high-dimensional energy landscapes to find low-energy configurations. This breakthrough has the potential to revolutionize fields such as protein folding and machine learning.
A new study reveals that the US Midwest has seen a significant increase in cover crop adoption, with 7.2% of cropland being planted with cover crops in 2021. This is attributed to government programs and funding initiatives, which have been shown to strongly correlate with the onset of cover crop assistance.
A team of researchers proposes an intelligent routing scheme to optimize link load balancing in software-defined networks, achieving significant performance improvements. Their algorithms outperform traditional methods, reducing maximum bandwidth by 24.6 percent in real-world topologies.
Scientists used AI-driven PandaOmics platform to analyze gene expression datasets from DNA repair diseases, identifying biomarkers associated with treatment response. The study focused on genes that stratify cancer patients by survival outcomes, providing potential targets for personalized therapies.
A study led by Kyoto University researchers found that AI-generated haiku poems, created without human intervention, were often indistinguishable from those penned by humans. In contrast, human-AI collaboration produced more creative works.
Researchers at MIT have developed a new architecture for optical neural networks, which can perform complex linear algebra operations using light signals. The new design eliminates uncorrectable errors that limited the scalability of earlier systems, enabling larger networks with improved accuracy.
A new algorithm unravels the structure of financial services on the Ethereum blockchain, revealing highly intertwined structures that involve risks not yet fully understood. The findings highlight the need for transparency and awareness among users, regulators, and policymakers to mitigate systemic risks associated with cryptoassets.
Researchers developed a new machine-learning framework that enables cooperative or competitive AI agents to consider the future behaviors of all agents, not just their teammates or competitors. This framework, FURTHER, uses two modules: an inference module and a reinforcement learning module, to enable agents to adapt their behaviors a...
Researchers found that algorithms classify white men better, reproduce false beliefs about physical attributes, and make stereotypical associations between men and sciences, women and arts. Under-representation of women in AI design and biased datasets contribute to the problem.
Chugh argues that AI risk assessments raise concerns about human biases and the absence of individualized decision-making in the judiciary. She advocates for more research on AI's role in the court system, prioritizing community-driven and individualized processes.
Researchers from the University of Johannesburg deployed Few Shot Learning (FSL) for NIALM, a non-intrusive appliance load monitoring system. FSL requires only 7 test images to recognize appliances with 97.83% accuracy, making it faster and more cost-effective than traditional Machine Learning.
Researchers have developed a new approach to phase retrieval in coherent X-ray imaging, using a complexity parameter to guide the algorithm. This methodology reduces artifacts and improves solution quality, resulting in higher-resolution images of micro- and nano-sized objects.
A recent study by UCF Associate Professor Varadraj Gurupur created an algorithm to predict and measure the incompleteness of electronic health records. The analysis found that missing information is a significant issue, with varying levels of incompleteness per year and no clear pattern of where it occurs.
A new AI tool helps governments decide whether to bail out a bank by predicting if the intervention will save money for taxpayers. The algorithm assesses financial implications and suggests optimal bailout strategies.
Researchers created a Facebook-like prototype platform, called Trustnet, where users rate posts as accurate or inaccurate before sharing. This approach showed that people effectively assess misinforming posts and share their assessments with others.
A novel multi-modal image retrieval system, DenseBert4Ret, has been developed by researchers from Gwangju Institute of Science and Technology (GIST) using deep learning algorithms. The system outperforms state-of-the-art models in retrieving images based on both image and text features.
A team of Lehigh University researchers is investigating the role of online media in shaping facts and influencing behavior during the COVID-19 pandemic. The project aims to better understand how individuals and communities develop their understanding of major events, ultimately acting on that understanding.
A researcher developed a system to determine the pose of cooperative spacecraft using ToF cameras and reflector markers. The system consists of three steps: reflector detection, matching, and pose calculation, and was tested in experiments with high accuracy.
Deep learning models can become less accurate in recognizing specific categories of images, sounds, or text after network pruning. Researchers demonstrate a technique to address this challenge, improving the fairness of deep learning models.
A research team developed a smart mask integrating an ultrathin soundwave sensor that detects breathing, coughing, and speaking sounds. The mask uses machine-learning algorithms to identify respiratory diseases and improve public health by enabling prolonged monitoring.
Researchers at MIT have developed a machine-learning model that captures how sounds propagate through spaces, allowing for accurate visual renderings of rooms. This technique has potential applications in virtual and augmented reality, as well as improving AI agents' understanding of their environment.
Researchers at FAU aim to empower amputees to maximize their individual potential for controlling the full dexterity of artificial hands using a novel bimodal skin sensor and machine learning algorithms. The project will develop customized prosthetic sockets and training programs to overcome limitations with current sensing technology.
A novel algorithm uses near-infrared spectroscopy to estimate intracranial pressure (ICP) based on hemoglobin levels. The research validates the accuracy of this method using invasive ICP data.
A deep learning algorithm trained on adult populations shows improved sensitivity for diagnosing thyroid nodules in children, with higher accuracy than traditional ACR TI-RADS. The study suggests the potential benefits of this new approach for evaluating thyroid nodules in younger patients.
Researchers at Max Planck Institute successfully revived ancient enzymes, revealing a novel protein component that increased CO2 specificity in Rubisco. This discovery provides new insights into the evolution of modern photosynthesis and suggests adding new components may improve its efficiency.
A University of Groningen team created two machine learning models to predict app removal risks, achieving accuracy rates of up to 79.2%. The models can help developers avoid bans and users protect their data.
Researchers at the University of Pennsylvania have developed an algorithm that enables 2D materials to maintain their mechanical strength after conversion into 3D structures. The algorithm is inspired by kirigami art and mimics the structure of nacre, a natural shell coating known for its robust mechanical properties.
A team of Illinois Tech researchers used machine learning to estimate the age and gender of individual users with high accuracy, raising questions about data security and privacy. The study highlights the need for better regulations and best practices to protect personal information from being misused.