Shulou(Shulou.com)11/24 Report--

CTOnews.com, Nov. 3 (Xinhua) recently, researchers have successfully developed a physical neural network that can learn and remember dynamically by imitating neural networks in the brain. The physical neural network consists of tiny nanowires that mimic synapses in the brain and perform tasks by responding to changes in electronic resistance at the intersection of the wires.

By identifying and calling the electrical pulse sequence, the physical neural network can use the dynamic data accessed online to perform real-time learning, image recognition and other tasks, avoiding the heavy use of memory and energy.

University of Sydney CTOnews.com Note: nanowire Network (Nanowire network) is a nanotechnology usually made of highly conductive silver wires invisible to the naked eye, covered with plastic materials and forming a network structure.

Each nanowire is about 1/1000 of the width of human hair, and together they form a random network that behaves much like the neural networks in our brains. They can self-assemble into a dynamic complex network with memory and processing capabilities, similar to the human brain. Now, an international team of researchers at the University of Sydney has shown that nanowire networks are not only similar to the human brain, but also can learn and remember like the human brain.

The physical neural network, similar to the human neural network, consists of thin wires with a diameter of 1/1000000000 meters, and processes information through a series of commands or algorithms that perform memory and learning tasks that respond to changes in electronic resistance at the intersection of nanowires, just like nodes in the "Pick-up Sticks" game.

Memory and learning tasks are implemented using simple algorithms that respond to changes in electronic resistance at the overlap of nanowires. This function, called a resistive memory switch, occurs when electrical input encounters a change in conductivity, similar to what happens to synapses in our brains.

Nanowire networks have learned to recognize handwritten numbers. This innovative technology not only saves energy, but also significantly reduces memory usage, paving the way for efficient, low-power machine intelligence that can handle complex real-world learning and memory tasks. Their groundbreaking research paper, published in Nature Communications, marks a major advance in the field of machine learning and artificial intelligence.

CTOnews.com encloses the address of the research paper: Zhu, R., Lilak, S.S., Loeffler, A. et al. Online dynamical learning and sequence memory with neuromorphic nanowire networks. Nat Commun 14, 6697 (2023). Https://doi.org/10.1038/s41467-023-42470-5

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.