Articles tagged with Algorithms
A new AI algorithm, APOLLO, accurately predicts microprocessor power consumption by analyzing just 100 signals out of millions, offering potential to improve efficiency and develop new processors. The technique has been validated on high-performance microprocessors and could help designers inform future chip design.
A doctoral student at Texas A&M University has discovered blood outgrowth endothelial cells (BOECs) as an alternative to induced pluripotent stem cells (IPSCs) for organs-on-chips, offering a cheaper and more accessible option for patient-specific research. The new cells can be isolated from just 50-100 milliliters of blood and have sh...
A recent study used computer vision algorithms to analyze nearly 9,400 Flickr photos taken along Colorado's Front Range, identifying preferred outdoor landscapes with moderate accuracy. The algorithm performed well for images of water, structures, and agricultural lands, but struggled with forests. Combining social media data with on-s...
A new machine learning-based algorithm can predict stable material compounds much faster than traditional methods, opening up new avenues for research and discovery. The researchers identified several thousand potential new compounds using the computer, offering a promising breakthrough in materials science.
Researchers developed a communication-effective, divide and conquer algorithm to address computational challenges in large-scale data analysis. The algorithm combines summary statistics from subsystems using confidence distributions, balancing statistical accuracy and computational efficiency.
A research team developed an AI framework that analyzes protein interactions to predict effective and low-toxicity cancer drug combinations. The framework, GraphSynergy, outperforms conventional models in identifying synergistic combinations.
A Michigan Tech-developed machine learning model uses probability to classify breast cancer shown in histopathology images and evaluate the uncertainty of its predictions. The model outperforms similar models and can measure uncertainty, promising time savings and referrals to human experts.
Researchers at Osaka City University developed a new quantum algorithm that calculates potential energy curves of molecules without controlled time evolutions. This addresses issues with conventional quantum phase estimation algorithms, enabling parallel processing and efficient full-CI calculations.
Researchers developed an algorithm predicting COVID-19 patient outcomes based on individual data, achieving high accuracy (over 90%) for up to ten days. This innovation enables hospitals to allocate staff and resources efficiently, potentially saving lives during future pandemic waves.
A KAUST team developed an improved method for detecting malicious intrusions using deep learning, achieving accuracy rates of up to 99% in simulations of different kinds of attacks. This stacked deep learning approach promises an effective defense against cyberattacks and could prevent outages in critical infrastructure.
A wearable device has been developed to detect and reverse opioid overdoses by injecting naloxone, a lifesaving antidote. The device, which senses when a person stops breathing and moving, has shown promising results in clinical trials.
A recent study published in The Journal of Finance and Data Science suggests that long-term returns are stable and easier to predict than short-term returns, which suffer from noise and 'look-ahead' bias. The study also found that larger training samples are required for optimal model performance
Machine learning enables better understanding of climate-induced hazards, predicting floods and landslides with high accuracy. The technology combines diverse data sources to assess risk extent, considering both triggering hazards and socio-economic vulnerability.
Researchers have developed a new method that uses deep neural networks to predict extreme heat waves with unprecedented accuracy, up to two weeks before they occur. This breakthrough has significant implications for risk management, planning, and warning systems, which will greatly improve public safety and support public policies.
A new study explores the problem of shortcuts in a popular machine learning method and proposes a solution that can prevent shortcuts by forcing the model to use more data. By removing simpler characteristics and asking the model to solve the task two ways, researchers reduce the tendency for shortcut solutions and boost performance.
Researchers developed an algorithm that predicts suicidal thoughts and behavior among adolescents with 91% accuracy, analyzing data from 179,384 students. The study reveals online harassment and bullying as leading predictors of suicidal ideation and behavior, with females more likely to experience suicidal thoughts.
Scientists have developed a software that adds missing sugar components to protein models created with AlphaFold, enabling more accurate structural predictions. This breakthrough has the potential to revolutionize workflows in biology, allowing scientists to understand proteins and their mutations faster than ever.
A new study improves AI diagnoses by penalizing algorithms for false negatives, which can be more urgent than accuracy. Researchers achieved significant improvements in precision and recall for chronic kidney disease and other conditions using cost sensitivity techniques.
A study by University of Minnesota Medical School researchers shows that merging AI with electrical brain stimulation can enhance specific brain functions related to self-control and mental flexibility. The method improved cognitive control in patients undergoing brain surgery for epilepsy, reducing anxiety and depression symptoms.
A new algorithm has been developed to train spiking neural networks, mimicking the human brain's structure and function. This approach enables these powerful, fast, and energy-efficient systems to solve complex tasks like image classification with high precision.
A new study at Columbia University Mailman School of Public Health uses machine learning to predict successful opioid dispensing models in U.S. counties. The analysis reveals that prescription drug monitoring program access provisions are the most consistent predictors of high-dispensing and high-dose dispensing counties.
Researchers develop an algorithm to find optimal or near-optimal solutions in the space of 'infeasible solutions' to speed up search, alleviating traffic congestion and improving city living. A novel solution to a combinatorial optimization problem in bicycle sharing systems is proposed.
A new project aims to empower people who are blind to independently review and protect their personal visual content from accidental privacy leaks. Researchers have developed novel computer vision algorithms that can detect sensitive information in images and videos, allowing users to blur or remove private content before sharing.
A team of researchers has developed a novel machine learning model that can identify medication orders requiring pharmacy intervention using provider behavior and contextual features. This approach reduces the risk of exposing sensitive patient data, while alleviating the workload of pharmacists and increasing patient safety.
Researchers at MIT develop a data-driven process using machine learning to optimize new 3D printing materials with multiple characteristics. The system lowers costs and lessens environmental impact by reducing chemical waste and suggesting unique chemical formulations that human intuition might miss.
Assistant Professor Kang Hao Cheong and his team discovered that chaotic switching for quantum coin Parrondo's games has similar underlying ideas to encryption. They found that using pre-generated chaotic sequences enhances the work, making it easier to invert the encrypted message to obtain the original state.
A new study by MIT researchers has found that blind and sighted readers have sharply different takes on what content is most useful to include in a chart caption. The study created a four-level framework for evaluating charts, which could help develop more effective tools for automatically generating captions and alternative text.
A new AI-powered algorithm, GEM, has been developed to quickly identify genetic causes of serious disease in newborns. The technology leverages machine learning and natural language processing to analyze vast amounts of genomic data and clinical records, achieving an accuracy rate of 92% compared to existing tools.
Researchers developed an algorithm that leverages medical informatics to predict autism spectrum disorder (ASD) diagnoses in young children. The new approach uses diagnostic codes from past doctor's visits to calculate a risk score, identifying which patients are at risk of receiving a confirmed ASD diagnosis.
Researchers at Tokyo University of Agriculture and Technology developed a simple and rapid method to detect amyloid protein in bovine livers using fluorescence fingerprint analysis. This approach allows for quick processing and accurate detection of AA amyloidosis, potentially leading to more efficient diagnostic tools for this disease.
A blockchain-based system allows leader robots to signal movements and add transactions to a chain, while malicious leaders forfeit tokens when caught in a lie. This limits the spread of incorrect information and enables follower robots to eventually reach their destination.
Researchers propose a solution using tethered unmanned aerial vehicles (TUAVs) to receive signals while minimizing uplink exposure. The system uses low-power 'green antennas' that only receive signals and do not radiate EMF, offering increased data transfer speeds.
By analyzing pitch, length, octaves, chords, dynamics, and main theme of four pieces from the mid-1800s Romantic era of classical music, researchers created protein songs with improved musicality. The study found that using a specific music style guided the structure of proteins to produce more pleasant melodies and harmonies.
SUTD researchers develop sensor that assigns dirt score to areas based on visual and tactile analysis, allowing for more efficient exploration of complex spaces. The sensor is integrated with a smart algorithm that directs the robot to focus on areas with high dirt probability.
Researchers developed a feature selection algorithm that uses boosting to select relevant features from high-dimensional data sets. The algorithm outperforms other methods in terms of accuracy and number of features used, making it more scalable and explainable.
A team of researchers, led by University of Houston associate professor Ryan Kennedy, has received a $750,000 NSF grant to create an algorithm-accountability benchmark. The project aims to establish general ways of analyzing algorithms and studying their impact on public policy decisions.
A multidisciplinary organization has reached consensus on guidelines for performing, interpreting, and reporting MR defecography. The consensus templates aim to standardize care for patients with evacuation disorders of the pelvic floor.
Researchers developed a new, accurate method to detect North Atlantic Right Whale up-calls using Multimodal Deep Learning algorithms. The technology outperformed conventional methods in detecting up-calls, non-up-calls, and false alarms.
A team of scientists from Incheon National University developed a programmable DNA-based microfluidic chip that can perform complex mathematical calculations, such as Boolean logic operations. The chip uses a motor-operated valve system to execute a series of reactions in rapid and convenient manner.
Researchers from the University of Cambridge have created a real-time approach to predict drone flight paths and intentions, enabling safer use of drones. The solution uses statistical techniques and radar data to identify potential threats before they enter restricted airspace.
A novel mortality risk prediction method helps tailor treatment decisions and transplant needs for patients based on individual symptoms. The new tool uses a random survival forest algorithm to predict individual mortality risk curves, calculate mortality at any given time, and provide a 95% confidence interval.
A study using machine-learning models trained on over 1 million companies reveals that AI can accurately predict the success of startups. The tool, developed by researchers, has the potential to help investors make informed decisions and avoid significant losses.
Researchers at The Hebrew University of Jerusalem have developed a new deep learning artificial infrastructure inspired by individual neurons. Their approach uses complex mathematical modeling to replicate the brain's electrical processes and create more intelligent AI systems.
A new electronic 'nose' has been developed to detect when a lung transplant is beginning to fail, with 86% accuracy. The device uses machine learning algorithms to analyze exhaled breath patterns and identify lung diseases, offering new hope for patients diagnosed with chronic allograft dysfunction.
Researchers developed an algorithm to rank apps based on their privacy scores, allowing users to easily find and install non-intrusive apps. The system considers two scores: permission and listener access, providing a ranking of apps from least intrusive to most private.
Researchers developed a new AI algorithm called 'basis profile curve identification' to simplify comparisons between effects of electrical stimulation on the brain. The algorithm may help understand which brain regions interact with each other, guiding placement of electrodes for treating network brain diseases.
A new algorithm, Phe2vec, accurately identified patients with certain diseases, outperforming traditional methods in classifying diagnoses. The study suggests that this automation will facilitate further research in clinical informatics.
Researchers developed a method to overlay a virtual scale on acquired endoscope images in real-time, allowing accurate estimation of colorectal polyp sizes. The approach uses triangulation principles and minimal image processing, enabling cost-effective diagnosis without adding extra instrumentation.
Researchers at Technical University of Munich have developed a new machine learning algorithm that can analyze complex markets and their equilibrium strategies. This breakthrough has potential applications in auction theory, wireless spectrum auctions, and more.
Researchers developed an AI tool that can quickly and accurately identify suspicious proteins in the body by analyzing their movements. The method, known as diffusional fingerprinting, uses machine learning algorithms to predict protein behavior with over 90% accuracy.
A new unsupervised machine learning algorithm, B-SOiD, developed by Carnegie Mellon University researchers makes studying animal behavior more accurate and efficient. The algorithm identifies patterns in an animal's body position to discover behaviors, removing human error and bias.
Researchers have developed an approach that predicts accurate structures computationally, overcoming the problem of determining molecular shapes. The algorithm succeeds even when learning from only a few known structures, making it applicable to difficult-to-determine molecules.
Researchers at UC San Diego developed a system that splits a single millimeter wave beam into multiple paths to improve reliability and throughput. The technology achieved high speeds of up to 800 Mbps with 100% reliability, even in outdoor tests over distances of 262 feet.
Researchers at Lawrence Berkeley National Laboratory have created a new mathematical algorithm to decipher the rotational dynamics of twisting particles in complex systems. By analyzing X-ray scattering patterns, they can gain insights into the function and properties of materials.
A team of researchers has created a new algorithm that determines the most efficient route for robots to navigate complex spaces. The RBF-Galerkin method combines two existing approaches to find the optimal solution, surpassing other methods in terms of cost and time efficiency.
Researchers used social media posts to predict COVID-19 case counts, achieving correlation rates of up to 0.98 and improving on existing Google Flu Trends algorithm results. The study provides a highly-adaptive approach for feature engineering in epidemic prediction.
Researchers at University of Michigan develop faster path planning approach for rubble-roving robots, enabling them to find stable paths in treacherous terrain more efficiently. The new algorithm outperformed traditional methods in success and total time to plan, with an 84% success rate in virtual experiments.
A new study uses machine-learning algorithms to predict the next phase of a traffic signal, giving bicyclists a smoother ride. The researchers achieved high accuracy with 85% prediction success rate, using LSTM and 1D CNN models.
Researchers at MIT have created an algorithm that enables drones to navigate complex obstacle courses at high speeds without crashing. The new approach combines simulations with real-world experiments, allowing drones to adapt to challenging aerodynamics and find the fastest routes.
Scientists at CiTIUS have developed a new fast support vector classifier (FSVC) that significantly improves data classification using Machine Learning techniques. The FSVC is much faster and operates with less memory than traditional approaches, making it suitable for large-scale classification problems.