A new study by New York University found that YouTube's recommendation algorithm prioritized election-fraud-related videos for users already skeptical about the 2020 presidential election's legitimacy. This highlights the dangers of opaque algorithms perpetuating misinformation and disinformation.
Researchers at Chalmers University of Technology developed a computer model to predict enzyme efficiency. This helps find efficient cell factories for producing biotech products like biofuels and medicines, and studies difficult diseases.
Xiu Yang, a 2022 NSF CAREER award recipient, is working on an algorithmic approach to model and overcome hardware errors in quantum computing. He aims to enable the technology to achieve its promise of unparalleled speed in solving complex problems.
A Brazilian research team has developed a novel method to sort specialty and standard coffee beans using multispectral imaging and machine learning. The technique, which does not require roasting or human intervention, uses images of the beans at different wavelengths to distinguish between quality levels.
Researchers designed a centered error entropy-based sigma-point Kalman Filter to enhance the filtering algorithm's robustness in spacecraft attitude determination. The proposed CEEUKF outperformed classical methods and other robust algorithms in simulating non-Gaussian noise, achieving higher accuracy and faster convergence rates.
Rice University's ROBE Array algorithm slashes the size of DLRM memory structures, allowing training on 100 megabytes of memory and a single GPU. The method matches state-of-the-art DLRM training methods with improved inference efficiency.
A team of Japanese researchers used reinforcement learning to study fluid mixing during laminar flow, achieving exponentially fast mixing without prior knowledge. The method also enabled effective transfer learning, reducing training time for new mixing problems, and has potential applications across various industries.
Researchers at the University of Oldenburg and Fraunhofer IWES collaborate on a new project to develop more accurate wind flow simulations using artificial intelligence. The goal is to reduce computing times and enhance precision, ultimately accelerating innovation in wind turbine design.
Researchers at MIT developed an AI model that can detect Parkinson's disease from breathing patterns, using a neural network to assess the presence and severity of the condition. The device is non-invasive and can be used in patients' homes without any bodily contact.
Researchers at NC State University have developed a cooperative distributed algorithm that allows autonomous vehicle software to make calculations more quickly, enabling real-time navigation of complex merging scenarios. The approach improves both traffic flow and safety, with zero incidents in simulations.
Scientists at Kyoto University developed two methods to identify RNA modifications impacting gene regulation and disease. Their approach uses probability algorithms with high-throughput sequencing technology, distinguishing pseudouridine substitutions from other base changes.
Bhattacharya's project uses topological abstraction to reduce complexity in robotic systems, enabling more efficient and accurate motion planning. The approach has potential applications in industries such as transportation, manufacturing, and healthcare.
Researchers at Princeton University used artificial intelligence to simulate ice formation by individual atoms and molecules with quantum accuracy. This breakthrough enables tracking of hundreds of thousands of atoms over longer timespans than previous simulations.
Researchers used natural language processing and machine learning to analyze nearly 35,500 death records, identifying the most common substances involved in overdose deaths. The system reduced data processing time by months, allowing for more rapid public health responses and interventions.
Researchers used machine learning algorithms and k-D tree data structure to identify 11 previously undetected space anomalies, seven of which are supernova candidates. The team analyzed digital images of the Northern sky taken in 2018 using a k-D tree to detect anomalies through the 'nearest neighbour' method.
Researchers optimized the ZZ SWAP network protocol, introducing a new technique to improve quantum error mitigation. This enables more efficient execution of quantum algorithms like QAOA, which can solve combinatorial optimization problems.
Researchers argue that using tools to estimate racial and ethnic information can identify algorithmic bias and combat disparities in healthcare. This approach can lead to more equitable pay-for-performance schemes and better clinical decision-making.
A team of Chan Zuckerberg Biohub scientists developed a deep-learning method, dubbed
A University of Washington team created a new tool that can design a 3D-printable passive gripper and calculate the best path to pick up an object. The designed grippers and paths were successful for 20 out of 22 objects tested, with two challenging shapes being the wedge and pyramid shape.
A new AI program identified four variables for a swinging double-pendulum, but the remaining two variables remain a mystery. The AI successfully predicted physical phenomena in other systems, such as air dancers and lava lamps, with varying numbers of variables.
A new AI algorithm, IcePic, has been developed to predict ice crystal formation with high accuracy. It outperformed human scientists in an online quiz, identifying areas where humans were wrong and providing valuable insights for atmospheric science research.
A new method can improve explosion detection by training computers to recognize synthetic infrasound signals, which reflect regional and global atmospheric changes. This approach broadens the usefulness of single-element infrasound microphones for detecting subtle explosion signals in near real-time.
Researchers at Max Planck Institute for Intelligent Systems created a robot dog named Morti that can walk smoothly within an hour. The robot uses a Bayesian optimization algorithm to learn from sensor data and adapts its virtual spinal cord, allowing it to optimize its walking pattern and minimize stumbling.
A study by Carnegie Mellon University researchers found that algorithmic transparency can have positive effects for firms, allowing them to motivate agents to improve valuable features. However, transparency may not always be beneficial for agents, as it can lead to a loss of predictive power and disadvantage high-type agents.
Researchers created a machine-learning algorithm to predict neighborhood racial segregation, showing 86% accuracy in one test. The map analyzed census data and showed less segregated areas becoming more mixed by 2030.
Researchers used logistic regression and a recommendation algorithm to predict CME arrival times, achieving better results than using either method alone. The hybrid model improved forecast accuracy by providing a reference for similar historical events.
Researchers developed a neural network algorithm that recognizes emotions and engagement from video images of faces, outperforming existing models in accuracy. The system can be integrated into video conferencing tools and online learning systems to analyze participant engagement and emotions.
A team of researchers led by Danilo Vasconcellos Vargas has developed a new method called 'Raw Zero-Shot' to evaluate the robustness of artificial neural networks in image recognition. The study found that Capsule Networks produced the densest clusters, indicating improved transferability and potential solutions for improving AI robust...
Researchers developed an automated method to create 3D images of leaked gas clouds, enabling precise location, volume, and concentration determination. This technology can provide early leak warnings, assess risk, or determine the best way to fix leaks in large facilities with stored toxic chemicals.
A new robotic system, FuseBot, has been developed to efficiently retrieve buried objects in piles. The system uses radio frequency signals and computer vision to reason about the probable location and orientation of objects under the pile, enabling it to find more hidden items than a state-of-the-art robotics system in half the time.
Researchers developed open-source software SHRY to find distinct substitution patterns in disordered systems, reducing computation time. The software uses group theory and canonical augmentation to efficiently analyze crystal structures with random substitutions.
A newly expanded data set of brain scans from stroke patients called ATLAS now includes 1,271 MRI images with manually segmented lesions, facilitating large-scale stroke recovery research. Researchers hope to develop algorithms to automate lesion segmentation, enabling clinicians to predict patient responses to therapies.
A novel algorithm, BLIND, enables robots to navigate through environments with obstacles by incorporating human feedback. Humans provide labels to refine the robot's trajectory, avoiding obstacles efficiently.
Researchers found that people are less morally outraged when gender discrimination occurs due to an algorithm rather than direct human involvement. The study's findings have broader implications for efforts to combat discrimination and may affect how companies are held liable.
SeqScreen, an open-source software toolkit, accurately characterizes short DNA sequences to detect pathogenic sequences. The program uses a curated database of thousands of gene sequences representing 32 types of virulence functions.
A team of scientists has developed a novel computational approach to analyze DNA sequences of thousands of bacteria, revealing previously unknown gene clusters responsible for producing metabolites of interest. The study highlights the potential of these bacterial compounds in treating colon cancer and improving treatment options.
A study by Thorsten Lehnert shows that corporate managers' behavior is linked to investor sentiment, predicting investment strategy success. The researcher found a significant relationship between market-level euphoria and investment factor performance, outperforming static strategies.
Researchers developed an algorithm to improve matching efficiency, considering user preferences and behavior. The new algorithm shows improved results in field experiments, with at least 27% more matches than the current one.
Researchers at MIT identified a flawed analysis of website-fingerprinting attacks and developed new techniques to prevent them. They found that attackers can use machine-learning algorithms to decode signals leaked between software programs, enabling them to obtain private information.
A new training algorithm for deep spiking neural networks (SNNs) uses biologically plausible spatiotemporal adjustment to improve performance and reduce energy consumption. This approach achieves competitive classification accuracy with only 3% of the energy used by traditional artificial neural networks.
A team of researchers from Waseda University developed a novel solution to efficiently solve complex optimization problems using Ising machines. Their hybrid algorithm reduces residual energy and reaches more optimal results in shorter time, increasing the machine's applicability across industries and sustainability practices.
A new AI method can distinguish between the overall sounds of healthy and unhealthy coral reefs, making it a valuable tool for monitoring reef health. The technique uses machine learning to analyze sound recordings and track the progress of reef restoration projects.
Researchers at Cornell University have developed a new algorithm for autonomous underwater sonar imaging that significantly improves speed and accuracy for identifying objects such as explosive mines and sunken ships. The new approach, called informative multi-view planning, integrates information about object locations with sonar proc...
MIT researchers develop an interactive design pipeline enabling users to create customized robotic hands with tactile sensors. The platform streamlines the process, allowing users to adjust palm and fingers and integrate tactile sensors, resulting in complex tasks like picking delicate items or using tools being performed flawlessly.
Researchers at Sainsbury Wellcome Centre found that mice can choose the best escape route after only 10 minutes of exploration, without needing to experience threat. The study suggests that mice use innate heuristics and natural exploration to learn this information.
Researchers used big data and machine learning methods to provide insight into equity valuation over the past two decades. However, a review by Professor Doron Nissim identified crucial areas that require more research attention.
Researchers have successfully processed sequences with a large neural network while consuming significantly less energy on neuromorphic hardware. This breakthrough showcases the potential of neuromorphic technology to improve the energy efficiency of AI workloads.
A machine learning algorithm has outperformed astronomers in analyzing microlensing data to find new exoplanets, revealing connections hidden in complex mathematics from general relativity. The AI algorithm uncovered a degeneracy that had been missed by experts, suggesting a broader theory is likely incomplete.
A new computer-aided diagnostic tool, Deep-Lung Parenchyma-Enhancing (DLPE), uses artificial intelligence to reveal signs of pulmonary fibrosis in COVID long-haulers, helping to explain respiratory symptoms and improve disease management.
A study by Thorsten Lehnert reveals that price noise from trading activities of mutual funds explains the success of 'bets against beta' strategies. These strategies, which involve taking short positions in assets with higher betas and long positions in those with lower betas, result in significant positive returns.
Researchers at MIT developed a technique that enables an autonomous vehicle to plot a provably safe trajectory in highly uncertain situations. The algorithm considers probability of observing different environmental conditions and obstacles, and formulates trajectory planning as a probabilistic optimization problem.
A new distributed learning technique, GD-SEC, reduces communication requirements in wireless architecture, improving efficiency and reducing computational cost. The method employs data compression to transmit only meaningful, usable data, enhancing the impact of machine learning while minimizing its limitations.
A new algorithm forecasts which hospitalized patients are at the highest risk of death from COVID-19, regardless of immune protection status or virus variant. The tool, called COVID-19 Disease Outcome Predictor (CODOP), uses blood measurements to predict prognosis and can help direct critical care resources.
Researchers at Duke University have developed a machine learning algorithm that incorporates known physics into neural networks, allowing for new insights into material properties and more efficient predictions. The approach helps the algorithm attain transparency and accuracy, even with limited training data.
A new nanosensor platform uses machine learning to analyze spectral signatures of carbon nanotubes for early detection of ovarian cancer. The approach detects biomarkers and recognizes the cancer itself, offering a promising alternative to traditional methods.
Harbir Antil leads a project to create scalable algorithms for optimizing nonlinear dynamical systems with uncertainty. The research aims to enable automated design, data analysis, and optimization using randomized preconditioning and compression methods.
A new observational study from Vanderbilt University Medical Center compares an artificial intelligence algorithm with face-to-face screening for estimating suicide risk. The ensemble learning method produced nearly always better predictions across the board compared to C-SSRS or VSAIL alone.
Researchers at North Carolina State University have developed FAXID, a hardware-based approach for detecting ransomware that is significantly faster than software-based methods. In proof-of-concept testing, FAXID demonstrated accuracy comparable to XGBoost but with speeds up to 65.8 times faster.
Researchers developed an algorithm to quickly identify two cancer drugs that work well together, reducing testing time from three years to eight weeks. The new method uses gene expression data to prioritize potential combinations and has been confirmed in lab tests.
Researchers create a new method for detecting diagnostic markers of epilepsy using EEG and MEG, enabling precise localisation of epileptogenic cortical structures. The biomimetic algorithm improves the effectiveness of neurosurgical interventions for epilepsy patients with resistant treatment.