Carding 4 Carders
Professional
When machines learn like the human brain.
The world is passionate about artificial intelligence (AI), which can process huge amounts of data. However, modern AI systems based on artificial neural networks consume a lot of energy, especially when working with real-time data.
Scientists have proposed a new approach to "machine intelligence". Instead of software for artificial neural networks, they developed a physical neural network at the hardware level, which works much more efficiently. These neural networks, created from silver nanowires, are able to learn how to recognize handwritten numbers in real time and memorize sequences of digits. The results of the study were published in the journal Nature Communications in collaboration with colleagues from the University of Sydney and the University of California.
With the help of nanotechnology, scientists have created networks of silver nanowires, the thickness of a thousandth of a human hair. These nanowires form a random network that resembles the structure of neurons in our brains. Such networks respond to electrical signals by changing the way electricity is transmitted at the intersection points of nanowires, which is similar to the operation of biological synapses.
The study shows that nanowire-based networks can be used for online learning. Unlike traditional machine learning, where data is processed in batches, the online approach feeds data to the system in a continuous stream. This method of on-the-fly learning, which is more efficient than traditional batch learning, requires less memory and energy.
In experiments, the nanowire-based network successfully coped with the tasks of recognizing and remembering numbers, demonstrating the potential for emulating brain-like learning and memory.
In general, research in the field of neuromorphic nanowire networks is just beginning, and new horizons of possibilities are opening up for scientists.
The world is passionate about artificial intelligence (AI), which can process huge amounts of data. However, modern AI systems based on artificial neural networks consume a lot of energy, especially when working with real-time data.
Scientists have proposed a new approach to "machine intelligence". Instead of software for artificial neural networks, they developed a physical neural network at the hardware level, which works much more efficiently. These neural networks, created from silver nanowires, are able to learn how to recognize handwritten numbers in real time and memorize sequences of digits. The results of the study were published in the journal Nature Communications in collaboration with colleagues from the University of Sydney and the University of California.
With the help of nanotechnology, scientists have created networks of silver nanowires, the thickness of a thousandth of a human hair. These nanowires form a random network that resembles the structure of neurons in our brains. Such networks respond to electrical signals by changing the way electricity is transmitted at the intersection points of nanowires, which is similar to the operation of biological synapses.
The study shows that nanowire-based networks can be used for online learning. Unlike traditional machine learning, where data is processed in batches, the online approach feeds data to the system in a continuous stream. This method of on-the-fly learning, which is more efficient than traditional batch learning, requires less memory and energy.
In experiments, the nanowire-based network successfully coped with the tasks of recognizing and remembering numbers, demonstrating the potential for emulating brain-like learning and memory.
In general, research in the field of neuromorphic nanowire networks is just beginning, and new horizons of possibilities are opening up for scientists.
