Showing 20 articles starting at article 361

< Previous 20 articles        Next 20 articles >

Categories: Mathematics: Modeling, Offbeat: Earth and Climate

Return to the site home page

     (via sciencedaily.com) 

The era of artificial-intelligence chatbots that seem to understand and use language the way we humans do has begun. Under the hood, these chatbots use large language models, a particular kind of neural network. But a new study shows that large language models remain vulnerable to mistaking nonsense for natural language. To a team of researchers, it's a flaw that might point toward ways to improve chatbot performance and help reveal how humans process language.

     (via sciencedaily.com) 

Scientists have confirmed that human brains are naturally wired to perform advanced calculations, much like a high-powered computer, to make sense of the world through a process known as Bayesian inference.

     (via sciencedaily.com) 

To address the lack of suitable training data for deep-learning semantic segmentation models in urban landscaping, researchers developed a method that generates a training dataset without the need for real images or a model of an existing city. The method, which is based on procedural modelling and image-to-image techniques, enables segmentation models to achieve comparable performance under some conditions at a fraction of the cost of real dataset generation.

     (via sciencedaily.com) 

Researchers have developed a new simulation model, which can predict flooding during an ongoing disaster more quickly and accurately than currently possible.

     (via sciencedaily.com) 

Researchers have focused on the moisture present in the air as a potential source of drinking water for drought-stressed populations. They reached a significant breakthrough in their efforts to create drinkable water out of thin air: a molecularly engineered hydrogel that can create clean water using just the energy from sunlight.

     (via sciencedaily.com)     Original source 

The sun and the sea -- both abundant and free -- are being harnessed in a unique project to create vertical sea farms floating on the ocean that can produce fresh water for drinking and agriculture.

     (via sciencedaily.com) 

A new artificial intelligence tool can help people with Parkinson's disease remotely assess the severity of their symptoms within minutes. While expert neurologists performed slightly better than the AI model, the AI model outperformed the primary care physicians with UPDRS certification.

     (via sciencedaily.com)     Original source 

Right at the bottom of the deep sea, the first very simple forms of life on earth probably emerged a long time ago. Today, the deep sea is known for its bizarre fauna. Intensive research is being conducted into how the number of species living on the sea floor have changed in the meantime. Some theories say that the ecosystems of the deep sea have emerged again and again after multiple mass extinctions and oceanic upheavals. Today's life in the deep sea would thus be comparatively young in the history of the Earth. But there is increasing evidence that parts of this world are much older than previously thought.

     (via sciencedaily.com) 

A team of scientists developed a new machine learning model for discovering critical-element-free permanent magnet materials based on the predicted Curie temperature of new material combinations.

     (via sciencedaily.com) 

Machine learning model provides quick method for determining the composition of solid chemical mixtures using only photographs of the sample.

     (via sciencedaily.com)     Original source 

Scientists have developed a fabric-based wearable device that 'taps' a user's wrist with pressurized air, silently helping them navigate to their destination. The study demonstrated that users correctly interpreted which direction the device was telling them to go an average of 87% of the time. Since the wearable embeds most of its control system within the fabric itself, using air instead of electronics, it can be built lighter and more compact than existing designs.

     (via sciencedaily.com)     Original source 

Researchers used advanced technology to study a massive aggregation of deep-sea octopus gathered at thermal springs near an extinct underwater volcano off the coast of Central California. Warm water from hydrothermal springs accelerates development of octopus embryos, giving young octopus a better chance of survival. The Octopus Garden is the largest known aggregation of octopus on the planet -- the size of this nursery, and the abundance of other marine life that thrives in this rich community, highlight the need to understand and protect the hotspots of life on the deep seafloor from threats like climate change and seabed mining.

     (via sciencedaily.com)     Original source 

Geoscientists have created a new method that can reconstruct the drift path and origin of debris from flight MH370, an aircraft that went missing over the Indian Ocean in 2014 with 239 passengers and crew. 

     (via sciencedaily.com)     Original source 

The sensory hairs of the Venus flytrap contain a heat sensor that warns the plant of bush fires. It reacts to rapid temperature jumps, as researchers have discovered.

     (via sciencedaily.com) 

Indoor air pollution may have met its match. Scientists have designed catalyst-coated lampshades that transform indoor air pollutants into harmless compounds. The lampshades work with halogen and incandescent light bulbs, and the team is extending the technology so it will also be compatible with LEDs.

     (via sciencedaily.com) 

Pouring flecks of rust into water usually makes it dirtier. But researchers have developed special iron oxide nanoparticles called 'smart rust' that actually makes it cleaner. The magnetic nanoparticles attract different pollutants by changing the particles' coating and are removed from water with a magnet. Now, the team is reporting a smart rust that traps estrogen hormones, which are potentially harmful to aquatic life.

     (via sciencedaily.com) 

Researchers hypothesize that a powerful type of AI model known as a transformer could be implemented in the brain through networks of neuron and astrocyte cells. The work could offer insights into how the brain works and help scientists understand why transformers are so effective at machine-learning tasks.

     (via sciencedaily.com) 

Researchers demonstrate how mathematical modeling combined with dynamic biomarkers can be used to characterize metastatic disease and identify appropriate therapeutic approaches to improve patient outcomes.

     (via sciencedaily.com)     Original source 

Imagine using insects as a source of chemicals to make plastics that can biodegrade later -- with the help of that very same type of bug. That concept is closer to reality than you might expect. Researchers will describe their progress to date, including the isolation and purification of the insect-derived chemicals and their conversion into their bioplastics.

     (via sciencedaily.com) 

A mathematical model provides new insights into the distribution of genetic information during bacterial cell division