Scientists from all over the world are working on various projects, whose purpose is to build computers with artificial intelligence, able to operate on the lines of work of the human brain. This is a very difficult task, but the latest discovery of specialists from the University of Southampton shows that it is not impossible.
Well, they have shown that resistors, which are able to remember their previous values of resistance, can be used to build a rich and advanced neural network. In the experiments, they prepared a network with metal oxides and have used them as artificial synapses.
In this way initiated the processes of learning without intervention from the outside, that is exactly how this is done in the human brain. But that’s not all, as created by researchers not only use less energy than their predecessors, but also remember your previous state, even when they are turned off (no charge).
So is a great material for development. “Internet of things” (Internet of Things), which is a smart system that will rise our interaction with different things of everyday life to a whole new, amazing level.
Of course, professionals are aware of the fact that the imitation of the work of the human brain with the help is still the song of the future, because it will need hundreds of billions, but they have shown that it is possible, and their achievements already can be used in the construction of smart objects.