Tech UPTechnologyThey create a supercomputer that mimics the human brain

They create a supercomputer that mimics the human brain

A group of scientists from the University of Manchester, in the United Kingdom, has launched, for the first time, the world’s largest neuromorphic supercomputer , designed and built to function in the same way as the brain of a human being. To do this, they have equipped it with a million processors.

This supercomputer has been baptized with the name of Spiking Neural Network Architecture –or, in its shortest form, SpiNNaker–, and is capable of completing more than 200 billion actions per second. This new machine has required an investment of 15 million pounds (about 17.2 million euros), twenty years to work on its conception and more than a decade for its manufacture , which began in 2006, according to the center. educational and research through a press release.

The SpiNNaker, built at the School of Computer Science at the University of Manchester, is capable of modeling more biological neurons in real time than any other machine on the face of the Earth . Biological neurons are basic brain cells present in the nervous system that communicate primarily by emitting spikes of pure electrochemical energy. Neuromorphic computing, the researchers explain, uses large-scale computer systems that contain electronic circuits to mimic these spikes in a machine.

Unlike conventional computers, which communicate by sending large amounts of information from point A to B over a standard network, SpiNNaker mimics the brain’s communication architecture to simultaneously send signals to thousands of destinations , behaving more as it does. the neural network of the brain.

A twist

Steve Furber, professor of Computer Engineering at the university and responsible for conceiving the idea of creating a computer with these characteristics, explains that “SpiNNaker completely rethinks the way conventional computers work. Basically, we have created a machine that works more like a brain than a traditional computer, which is extremely exciting. “

Since April 2016, SpiNNaker has been simulating neuron activity using half a million core processors, but the machine now has twice that capacity , Furber says. With the support of the Human Brain Project (HBP), a medical-scientific and technological project funded by the European Union, SpiNNaker will continue to allow scientists to create detailed brain models.

Researchers have set a goal to model up to one billion biological neurons in real time, and they are close to doing so. But what does that imply? Is SpiNNaker close to behaving like a real human brain? The answer is no. To give us an idea of the scale, they explain that in a mouse brain we find around 100 million neurons, while in the human the figure is a thousand times higher (that is, 100,000 million neurons), which still we would be far away.

So what can they use this supercomputer that works like a brain for? Among its main applications will be to help neuroscientists better understand how our own brain works – and, therefore, neurological diseases such as Parkinson’s – and it achieves this by performing real-time simulations that would not be possible to carry out with other machines. .

Image: University of Manchester

They design a 'memory prosthesis' to restore memories

These are electrodes that are inserted into the brain and mimic what the hippocampus does naturally. It works better in people with poor memory.

A small genetic mutation made us smarter than Neanderthals

This is what made us so different from Neanderthals. A subtle change in a single gene gave us the advantage.

The human brain did not shrink 3,000 years ago

A recent study shatters the theory of evolution. The work bucks the widely held belief that modern humans experienced an evolutionary decline in brain size. According to experts, the size of the human brain has not changed in probably 300,000 years.

The mysterious world of phobias

Phobias are a type of psychological alteration where the causes that trigger the crises (the moments in which their symptoms are most noticeable) are very diverse and, in some cases, curious or simply strange.

They create an artificial intelligence capable of thinking like a baby

The AI has been trained with videos designed to represent the same basic knowledge that babies have in their first months of life.