From neurons to the chip
Today one of the greatest challenges for engineers is the realization of machines that go beyond the dominant model so far where the execution of tasks is made from precise instructions.
More than 50 years ago the mathematician John Von Neumann wrote the following phrase that currently makes a lot of sense and is oriented to the development of intelligent machines "a deeper study of the nervous system could transform our understanding of mathematics and logic."
This new direction involves the development of machines endowed with an intelligent artificial mind, that is to say, less sequential and more creative and intuitive in order to perform more sophisticated functions such as: reasoning, planning and having self-awareness. Therefore, seek to imitate the human brain.
But how to achieve it?
In recent years some laboratories have worked on the development of "neuromorphic chip" inspired by the cerebral cortex. These prototypes are small pieces of silicon made up of a few hundred artificial neurons and a memory block that establishes the property of the synapses that can be made between the different neurons.
With this system, a computer would have the capacity to manage many more processes simultaneously than a normal computer, in addition to receiving stimuli from abroad and integrate them without needing to be reprogrammed to develop a new task.
According to the studies and the projections made on this technology, around the year 2045 the machines will reach a level like the human brain to perform their tasks. Although the goal is still far from being achieved, it is really exciting to think that it can be achieved.