Showing 20 articles starting at article 21
< Previous 20 articles Next 20 articles >
Categories: Geoscience: Earthquakes, Mathematics: Modeling
Published Method prevents an AI model from being overconfident about wrong answers
(via sciencedaily.com) Original source Thermometer, a new calibration technique tailored for large language models, can prevent LLMs from being overconfident or underconfident about their predictions. The technique aims to help users know when a model should be trusted.
Published Demographics of north African human populations unravelled using genomic data and artificial intelligence
(via sciencedaily.com) Original source A new study places the origin of the Imazighen in the Epipaleolithic, more than twenty thousand years ago. The research concludes that the genetic origin of the current Arab population of north Africa is far more recent than previously believed, placing it in the seventh century AD. The team has designed an innovative demographic model that uses artificial intelligence to analyze the complete genomes of the two populations.
Published Towards smart cities: Predicting soil liquefaction risk using artificial intelligence
(via sciencedaily.com) Original source Soil liquefaction that results in infrastructure damage has long been a point of contention for urban planners and engineers. Accurately predicting the soil liquefaction risk of a region could help overcome this challenge. Accordingly, researchers applied artificial intelligence to generate soil liquefaction risk maps, superseding already published risk maps.
Published Researchers explore the potential of clean energy markets as a hedging tool
(via sciencedaily.com) Original source Clean energy investments offer potential stability and growth, especially during volatile market conditions. A recent study explored the relationship between clean energy markets and global stock markets. Significant spillovers were observed from major indices like the SP500 to markets such as Japan's Nikkei225 and Global Clean Energy Index. These interactions suggest opportunities for optimizing investment portfolios and leveraging clean energy assets as hedging tools in volatile market environments.
Published Breaking MAD: Generative AI could break the internet, researchers find
(via sciencedaily.com) Original source Researchers have found that training successive generations of generative artificial intelligence models on synthetic data gives rise to self-consuming feedback loops.
Published When allocating scarce resources with AI, randomization can improve fairness
(via sciencedaily.com) Original source Researchers argue that, in some situations where machine-learning models are used to allocate scarce resources or opportunities, randomizing decisions in a structured way may lead to fairer outcomes.
Published Raindrops grow with turbulence in clouds
(via sciencedaily.com) Original source Tackling a long-time mystery, scientists have found that the turbulent movements of air in clouds play a key role in the growth of water droplets and the initiation of rain. The research can improve computer model simulations of weather and climate and ultimately lead to better forecasts.
Published Can a computer tell patients how their multiple sclerosis will progress?
(via sciencedaily.com) Original source Machine learning models can reliably inform clinicians about the disability progression of multiple sclerosis, according to a new study published this week in the open-access journal PLOS Digital Health by Edward De Brouwer of KU Leuven, Belgium, and colleagues.
Published Large language models don't behave like people, even though we may expect them to
(via sciencedaily.com) Original source People generalize to form beliefs about a large language model's performance based on what they've seen from past interactions. When an LLM is misaligned with a person's beliefs, even an extremely capable model may fail unexpectedly when deployed in a real-world situation.
Published Risks and benefits of integrating AI into medical decision-making
(via sciencedaily.com) Original source Researchers found that an artificial intelligence (AI) model solved medical quiz questions -- designed to test health professionals' ability to diagnose patients based on clinical images and a brief text summary -- with high accuracy. However, physician-graders found the AI model made mistakes when describing images and explaining how its decision-making led to the correct answer.
Published There is mathematical proof in the pudding
(via sciencedaily.com) Original source In blockchain development, there is a rule of thumb that only two of scalability, security, and decentralization are valid simultaneously. However, the mathematical expression of that rule was still a work in progress. Researchers discovered a mathematical expression for the blockchain trilemma. In the formula for Proof of Work-based blockchains, including Bitcoin, the product of the three terms--scalability, security, and decentralization--is 1.
Published Are AI-chatbots suitable for hospitals?
(via sciencedaily.com) Original source Large language models may pass medical exams with flying colors but using them for diagnoses would currently be grossly negligent. Medical chatbots make hasty diagnoses, do not adhere to guidelines, and would put patients' lives at risk. A team has systematically investigated whether this form of artificial intelligence (AI) would be suitable for everyday clinical practice. Despite the current shortcomings, the researchers see potential in the technology. They have published a method that can be used to test the reliability of future medical chatbots.
Published Scientists use AI to predict a wildfire's next move
(via sciencedaily.com) Original source Researchers have developed a new model that combines generative AI and satellite data to accurately forecast wildfire spread.
Published How to assess a general-purpose AI model's reliability before it's deployed
(via sciencedaily.com) Original source A new technique estimates the reliability of a self-supervised foundation model, like those that power ChatGPT, without the need to know what task that model will be deployed on later.
Published A new neural network makes decisions like a human would
(via sciencedaily.com) Original source Researchers are training neural networks to make decisions more like humans would. This science of human decision-making is only just being applied to machine learning, but developing a neural network even closer to the actual human brain may make it more reliable, according to the researchers.
Published When to trust an AI model
(via sciencedaily.com) Original source A new technique enables huge machine-learning models to efficiently generate more accurate quantifications of their uncertainty about certain predictions. This could help practitioners determine whether to trust the model when it is deployed in real-world settings.
Published Artificial intelligence speeds up heart scans, saving doctors' time, and could lead to better treatment for heart conditions
(via sciencedaily.com) Original source Researchers have developed a groundbreaking method for analysing heart MRI scans with the help of artificial intelligence, which could save valuable time and resources, as well as improve care for patients.
Published Researchers customize AI tools for digital pathology
(via sciencedaily.com) Original source Scientists developed and tested new artificial intelligence (AI) tools tailored to digital pathology--a rapidly growing field that uses high-resolution digital images created from tissue samples to help diagnose disease and guide treatment.
Published Researchers introduce generative AI for databases
(via sciencedaily.com) Original source Researchers have developed an easy-to-use tool that enables someone to perform complicated statistical analyses on tabular data using just a few keystrokes. Their method combines probabilistic AI models with the programming language SQL to provide faster and more accurate results than other methods.
Published Innovative, highly accurate AI model can estimate lung function just by using chest x-rays
(via sciencedaily.com) Original source An artificial intelligence (AI) model that can estimate with high accuracy a person's lung function just by using a chest radiograph has been successfully developed.