Advertisement

High-tech brain is scientists' goal

A nanocomponent that is capable of learning: The Bielefeld memristor built into a chip here is 600 times thinner than a human hair. Credit: Bielefeld University
A nanocomponent that is capable of learning: The Bielefeld memristor built into a chip here is 600 times thinner than a human hair. Credit: Bielefeld University

BIELEFELD, Germany, Feb. 26 (UPI) -- German scientists said they're making headway toward creating a computer that would work like a human brain, learning by itself without programming.

Researchers at Bielefeld University say they are experimenting with memristors -- electronic microcomponents that imitate natural nerves.

Advertisement

After constructing a memristor that is capable of learning, Andy Thomas from the university's Faculty of Physics said he is using them as key components in a blueprint for an artificial brain.

Memristors are the electronic equivalent of the brain's synapses, the bridges across which nerve cells (neurons) communicate with each other and which increase in strength the more often they are used.

Like synapses, memristors can learn from earlier impulses, Thomas said in a university release Tuesday.

Because of their similarity to synapses, memristors are particularly suitable for building an artificial brain, he said.

"They allow us to construct extremely energy-efficient and robust processors that are able to learn by themselves," Thomas said, suggesting principles taken from nature can be recreated in technological systems to develop a neuromorphic or nerve-like computer

"This is all possible because a memristor can store information more precisely than the bits on which previous computer processors have been based," Thomas said. "This is how memristors deliver a basis for the gradual learning and forgetting of an artificial brain."

Advertisement

Latest Headlines