It is a breakthrough in artificial intelligence able to recognize numbers spoken by different individuals with 99.6% accuracy.
Researchers at the Joint Center for Physics CNRS / Thales , the Nanosciences and Nanotechnologies Center (CNRS / Université Paris Sud), in collaboration with US and Japanese researchers developed the first nano-artificial neuron capable of recognizing those numbers by different individuals. This nano-electronic neuron is a breakthrough in artificial intelligence and their potential applications .
The latest artificial intelligence algorithms are able to recognize visual cues and vocal with high levels of performance. But these programs run on conventional computers used 10,000 times more energy than the human brain .
For this reason it is necessary to reduce electricity consumption and thus a new type of computer is needed. This is inspired by the human brain and comprises a large number of neurons and synapses miniaturized. So far, however, it had not been possible to produce an artificial nano-neuron sufficiently stable to process information reliably.
Today for the first time, researchers have developed a nano-neuron with the ability to recognize numbers spoken by different individuals with 99.6% accuracy . This development was based on the use of an exceptionally stable magnetic oscillator.
Each turn of this nano-compass generates an electrical output, which effectively mimics the electrical impulses produced by biological neurons. In the coming years, these magnetic nano-neurons could be interconnected through artificial synapses, as newly developed for the analysis and classification in real time big data.
The project is a collaborative effort between basic research laboratories and applied research partners. The long - term goal is to produce chips miniaturized extremely energy efficient with the need to learn and adapt to ever - changing and ambiguous intelligence real - world situations. These chips electronics have many practical applications such as providing an intelligent guide robots or autonomous vehicles, help doctors in their diagnosis and improve medical prostheses.
Source: PCWorld, William Peña
