Research using machine learning to solve scientific problems. Covers AI quality control, sea-surface temperature estimation, label-free cell analysis, and cyber defense.
The Buck Institute and Phenome Health have been awarded up to $52M by ARPA-H to develop a groundbreaking research project that aims to predict and prevent diseases using advanced analytics and AI. The project, known as PATH, will utilize machine learning and digital wearables to create personalized recommendations for healthy aging.
Researchers developed two machine learning algorithms to determine whisky origin and identify strongest aromas, outperforming human experts. The algorithms accurately classified whiskies into American or Scotch categories with over 90% accuracy.
Robert Johansson's Machine Psychology concept combines adaptive artificial intelligence with psychological learning principles to create a more intelligent AI system. The goal is to implement human-like intelligence in machines, enabling them to learn from experiences and apply knowledge across various situations.
Physicists from the University of Konstanz have created a solution using microrobots and counterfactual rewards to ensure fair distribution of load in collective tasks. The approach enhances efficiency and provides insights into improving teamwork in various collective systems.
Researchers at KAIST developed a new method to learn without weight transport, enabling faster and more accurate learning. By pre-training with random noise, the team showed that neural networks can achieve high learning efficiency and solve the weight transport problem.
The study found that individuals with accelerated biological ageing had poorer health outcomes, while those with decelerated ageing had weaker links to good health. Metabolomic ageing clocks have the potential to identify early signs of declining health and inform preventative strategies.
A new study by UCL researchers found that AI systems amplify human biases, leading to a snowball effect where small initial biases increase the risk of human error. The researchers demonstrated real-world consequences, including overestimating white men's likelihood of holding high-status jobs and underestimating women's performance.
Researchers developed a new benchmark for health care using reinforcement learning, which shows promise in managing chronic or psychiatric diseases. However, current methods are data-hungry and fail to perform accurately when tested on real-world data.
Researchers at MIT have developed Boltz-1, an open-source AI model that achieves state-of-the-art performance in predicting biomolecular structures. The model surpasses AlphaFold3, which is limited to academic research and commercial use, by incorporating new algorithms and improving prediction efficiency.
A new tool developed by Penn State researchers uses computer vision and artificial intelligence to analyze placenta images, detecting abnormalities and risks such as neonatal sepsis. The PlacentaCLIP+ model has the potential to transform neonatal and maternal care in low- and high-resource settings.
Researchers developed a unique numerical decision-making framework for solar panel protection against extreme weather conditions. The framework treats individual panels as independent decision-makers, identifying creative solutions to reduce stress and minimize damage during high-wind events.
Researchers achieved near-perfect accuracy in detecting Parkinson's disease by analyzing brain responses to emotional situations. The study identified distinct patterns in how patients processed emotions, enabling accurate differentiation between patients and healthy controls.
A multimodal machine learning model outperformed clinical and genomic models in predicting outcomes for HR-positive, HER2-negative breast cancer patients receiving CDK4/6 inhibitor combinations. The model integrated clinical and genomic factors to identify high-risk patients with a 6.5-fold difference in hazard ratio.
Researchers at Graz University of Technology developed a new database to improve speech recognition of Austrian German using speech data from 38 speakers. They found that traditional HMM-based systems are more robust for short sentences and dialectal language, while transformer-based models excel with longer sentences and context.
Insilico Medicine has received its first clinical milestone payment of $10 million from Exelixis for XL309, a selective USP1 inhibitor discovered with the company's AI platform. The drug is being developed for advanced solid tumors and has shown efficacy in preclinical studies.
A team of researchers has developed a novel technique to steal artificial intelligence (AI) models by monitoring electromagnetic signals. The method allows attackers to recreate the high-level features of an AI model with 99.91% accuracy, potentially undermining intellectual property rights and exposing sensitive data.
Researchers developed a deep learning model that classifies pancreatic cancer into molecular subtypes using histopathology images, achieving high accuracy and rapid turnaround time. The AI tool has the potential to improve patient outcomes by enabling timely and tailored treatment strategies.
Researchers at Drexel University developed an AI tool using large language models to identify and suggest alternative words that stigmatize people with substance use disorder. This framework aims to educate users and preserve supportive spaces in online forums.
A new technique identifies and removes specific points in a training dataset that contribute most to a model's failures on minority subgroups. This approach maintains the overall accuracy of the model while improving its performance regarding underrepresented groups.
Silvia Blemker, a University of Virginia biomedical engineer, has been elected Fellow of the National Academy of Inventors (NAI) for her work on muscle health. Her patented technology, Image-based Identification of Muscle Abnormalities, uses advanced imaging and analytics to provide detailed insights into muscle health.
Researchers at Pusan National University developed a hybrid model to predict metal wear in magnesium alloys, enabling safer, lighter designs. The model combines machine learning and physics to improve fatigue life prediction, offering greater predictive reliability for enhanced safety and longevity.
A groundbreaking AI model called NitroFusion creates images in seconds using modest hardware, eliminating the need for large computing resources. The open-source technology enables creative professionals and individuals to produce high-quality images affordably.
Researchers from Bar-Ilan University discover that classifying objects together through Multi-Label Classification can yield better results than detecting individual objects. This new method allows networks to learn correlations between object combinations, making them more recognizable in real-life applications such as autonomous vehi...
Researchers used machine learning to predict multiple types of intelligence from brain connections, with general intelligence performing best. The model's accuracy improved when trained on theory-driven connections, suggesting there are still unknown aspects of intelligence to discover.
A new multi-target quantum compilation algorithm developed by Tohoku University's Dr. Le Bin Ho improves the flexibility and performance of quantum computers. This allows for efficient handling of complex systems and tasks involving multiple variables in quantum machine learning.
A new research paper published in Oncotarget introduces an innovative AI tool combining CT scans and body composition data to predict severe liver problems in primary sclerosing cholangitis (PSC) patients. The model achieved impressive results, correctly identifying at-risk patients with 97% accuracy.
Physicists at the University of Michigan have developed an algorithm that enables materials to learn and adapt, mimicking brain-like behaviors. This breakthrough has implications for the development of advanced materials with self-tuning properties.
Researchers at Tokyo University of Science have developed a new method called black-box forgetting, which enables selective removal of unnecessary information from large pre-trained AI models. This approach enhances model efficiency and improves privacy by reducing computational resources and information leakage.
Researchers developed a deep-learning model to predict organoid development at an early stage, outperforming human experts in accuracy and speed. The model classifies bright-field images of organoids into three quality categories, indicating their potential for regenerative medicine applications.
A new grant will support a five-year investigation into the role of the brain's reward network in two neurodevelopmental conditions. Researchers will use high-resolution fMRI scans to analyze how the reward network interacts with other parts of the brain in autism, ADHD, and co-occurring autism and ADHD.
Researchers used a machine-learning technique to accelerate discovery of materials for film capacitors, identifying a compound with record-breaking performance. The study aims to improve capacitor shielding properties and enhance energy savings in common electric power applications.
A team of researchers has successfully identified a compound with record-breaking performance in film capacitors, a crucial component in renewable energy technologies. The breakthrough was achieved using a machine-learning technique that screened nearly 50,000 chemical structures to find the optimal material.
The US Naval Research Laboratory (NRL) will present its latest advancements in Earth and space sciences at the American Geophysical Union (AGU) Conference. NRL researchers will share their work on topics such as atmospheric data assimilation, ocean sciences, and geostationary ocean color.
A new study improves breast cancer risk prediction by analyzing up to three years of previous mammograms, identifying individuals at high risk 2.3 times more accurately than the standard method. The AI method considers subtle changes in mammogram images and holds up well across diverse settings.
Researchers at Seoul National University have developed a machine learning-based design of experiments method that optimizes the performance and process conditions of organic thermoelectric devices, enabling efficient evaluation of key variables and reducing experimental time. The technology has the potential to significantly improve d...
Dr Sanja Panovska has been awarded an ERC Consolidator Grant for her EXCURSION project, which aims to develop the first data-based model of the Earth's magnetic field over the last 780,000 years. The project will investigate geomagnetic excursions and their impact on technological infrastructure, environment, and climate.
Researchers analyzed 81 common household items for chemical makeup and exposure risks. The study used advanced chromatography and machine learning methods to identify chemicals that could pose negative health effects, such as synthetic antioxidant BKF, when exposure reached a certain threshold.
Dr. Sebastian Stich aims to create more efficient and adaptable machine learning models using collaborative learning approaches. The goal is to reduce computational power demands and costs, making it accessible to smaller players in fields like medicine.
The book examines AI's current advances, hurdles, and potential, emphasizing the need for science to maintain core norms and values. Experts advocate for human accountability and responsibility when using AI in research, highlighting the importance of transparent disclosure and attribution.
Researchers developed a platform to help AI learn complex tasks through nuance and real-time instruction, achieving up to a 30% increase in success rates. The GUIDE framework allows humans to provide ongoing, nuanced feedback, fostering incremental improvements and deeper understanding.
The Polymathic AI team has released two massive datasets for training artificial intelligence models to find and exploit transferable knowledge between seemingly disparate fields. The datasets include data from dozens of sources, covering astrophysics, biology, acoustics, chemistry, fluid dynamics, and more.
Researchers have uncovered key insights about how liquid crystals transform between different phases using direct simulation and machine learning. This study provides a clearer understanding of the microscopic-level changes in these materials, which could lead to new possibilities for advanced materials development.
A new AI model, Spherical DYffusion, can predict climate patterns over 100 years in just 25 hours, making it 25 times faster than the state of the art. The model uses generative AI and physics data to achieve this breakthrough, reducing the need for supercomputers.
A recent study presents a machine learning technique for rapid CO retrieval from hyperspectral satellite data, providing insights into air quality and pollutant transport over East Asia. The approach outperforms traditional physical methods in terms of consistency and accuracy.
A machine learning algorithm developed by researchers at Aarhus University can identify patients at high risk of involuntary admission within six months. The study, based on electronic health record data, suggests that the algorithm can improve targeted treatment and prevention in psychiatric services.
Insilico Medicine has developed an innovative platform using generative AI to discover novel molecules. The company has nominated over 20 preclinical candidates and received IND clearance for 10 molecules.
A new study found that brain stimulation effectiveness is linked to individual learning abilities, not age. The study used atDCS on participants with optimal and suboptimal learning strategies, revealing accelerated accuracy improvement in suboptimal learners.
NCSA launches Harbor, a fast storage service for global home and software directories, boosting Delta system performance by 400%. Consolidating file systems to VAST namespace improved uptime, load times, and user experience. With Harbor and VAST Data Platform, NCSA achieves scalable and efficient AI infrastructure.
Recent Nobel Prizes in physics and chemistry have recognized the convergence of AI with physics and chemistry, emphasizing the need for interdisciplinary research. Researchers advocate for nurturing AI-enabled polymaths to bridge the gap between theoretical advancements and practical applications.
University of Toronto researchers developed a new computational method to study genome organization, uncovering previously unknown patterns in human chromosomes. The method uses machine learning to analyze high-throughput chromosome conformation capture data and sheds light on chromosomal re-organization leading to disease development.
A new AI tool generates realistic satellite images of future flooding, which can help communities visualize and prepare for approaching storms. The method combines a generative artificial intelligence model with a physics-based flood model, producing more accurate and realistic images than an AI-only approach.
A study found that ChatGPT provides higher fatality numbers when asked in Arabic compared to questions in Hebrew, highlighting the impact of user language on information dissemination. The researchers believe this has profound social implications, as it can shape perceptions of conflict and fuel biases.
The HumanTech Summit 2024 conference brings together experts in technological and social innovation, discussing key topics such as data security and interpersonal relationships. With a proven legacy of impact, the event has attracted over 1,400 participants and featured renowned keynote speakers from top universities.
The Mount Sinai Health System has received the 2024 CHIME Digital Health Most Wired recognition as a certified level 9 in acute care and level 8 in ambulatory care. The system leverages digital health solutions to enhance patient care and reduce administrative burden.
Researchers at Cold Spring Harbor Laboratory have devised a potential solution to the paradox of animal innate abilities using artificial intelligence. The genomic bottleneck algorithm allows for compression levels unseen in AI, enabling faster runtimes and potentially leading to more evolved AI systems.
The São Paulo Advanced School on Disordered Systems will bring together students and researchers in complexity, bio-inspired applications, information science, and quantum materials. The school, supported by FAPESP, aims to establish a common forum for learning and discussing theories of general interest.
Researchers successfully detect neutron participating in DVCS reaction using a new detector installed at Thomas Jefferson National Accelerator Facility. The experiment provides unprecedented insight into the distribution of partons inside neutrons, a crucial step towards understanding nucleon structure and spin.
Researchers at MIT have introduced a new algorithm that strategically selects the best tasks for training an AI agent, resulting in improved performance and reduced training costs. The technique outperforms existing methods by five to 50 times, making it more efficient and effective.
A new study from Regenstrief Institute presents a low-cost, scalable methodology for early dementia risk identification. The technique uses machine learning to extract relevant phrases from medical notes, providing healthcare providers with an individualized dementia risk prediction or evidence of mild cognitive impairment.
The Margot and Tom Pritzker Prize for AI in Science Research Excellence recognizes exceptional contributions to artificial intelligence and scientific inquiry. Awardees will receive $50,000 and be honored at the University of Chicago and Caltech Conference on AI+Science.