Neuromorphic devices and machine learning combine to make brain-like devices possible
Peer-Reviewed Publication
Updates every hour. Last Updated: 8-Nov-2025 03:11 ET (8-Nov-2025 08:11 GMT/UTC)
Here, researchers from Beijing Institute of Nanoenergy and Nanosystems (Chinese Academy of Sciences) and Yonsei University present the latest progress in neuromorphic computing by integrating various neural networks, including SVM, ANN, CNN, RNN, and RC. Starting from the structure of synapses and neurons, they explore how these networks can be combined with neuromorphic devices to replicate more complex brain-like computations. They also propose future development directions for neuromorphic devices, focusing on advancements in their structures, materials, and applications across diverse fields such as vision, touch, hearing, smell, pain and other senses.
Omega-3 fatty acids are known to be an essential part of a healthy diet. As humans cannot produce them, they have to be consumed in sufficient amounts. However, omega-6, -7, -9, and -10 fatty acids also play important roles in the metabolism of fats. These numbers indicate the position of the first double bond in a fatty acid chain. Deviations in the omega position can signal enzyme malfunctions or pathological metabolic processes, such as those occurring in cancer. Now, researchers at the University of Graz and the University of California, San Diego present in Nature Communications a novel, effective method to determine omega positions of lipids – the scientific term for fats – in complex biological samples including human tissues and blood.
A research team has introduced DepthCropSeg, a powerful, depth-informed crop segmentation approach that achieves near-supervised performance without the need for pixel-level manual annotations.