Search
Researchers trained artificial intelligence models on roughly 190,000 images, achieving over 99% accuracy when identifying roasted coffee varieties.
A new deep learning model predicts airplane hard landings from flight recorder data while clearly showing which moments and instruments triggered the alert.
Scientists tested different computer models to predict severe storm days in Ranchi, India, finding that boosting methods spot dangerous weather most reliably.
A newly designed system pairs a specialised language model with reinforcement learning to spot complex website security flaws that regular scanners miss. By practising over time, the tool learned to link subtle weaknesses together like moves on a chessboard.
Researchers found that a compact model called MobileNetV2 reliably handles dog breed identification entirely offline on a smartphone in under a tenth of a second.
Researchers reviewing medical imaging find that neural networks can process raw sound echoes into high-resolution scans at ultrafast speeds, bypassing slow mathematical calculations.
A virtual computer model uses continuous sensor data to forecast dangerous glucose shifts in septic intensive care patients 15 to 30 minutes in advance, updating in seconds on ordinary hardware.
A computer framework called CMS-NAS automatically designs tiny, accurate neural networks to predict epileptic seizures from brain activity. It combines image filters, memory models, and brain-like spiking neurons to run on low-power wearable chips.
Treating waste shells from Calophyllum inophyllum nuts with common chemicals turns them into porous carbon capable of storing rapid bursts of electrical charge.
Marine clouds do not react to pollution straight away; their cover peaks a day later and hundreds of kilometres downwind. This delay reshapes how scientists calculate cloud cooling effects.