Scientists are ready to launch a supercomputer that will simulate the human brain: the date has been announced

In Australia, a group of scientists is preparing to launch the DeepSouth supercomputer in 2024, which is capable of fully simulating the synapses of the human brain. It is hoped that this machine will help us understand how our brains process huge amounts of information while consuming relatively little energy.

According to New Scientist, researchers at the International Center for Neuromorphic Systems (ICNS) in Sydney, Australia, are working on the technological development. Two of the world's largest computer technology manufacturers, Intel and Dell, are partners in the project.

Unlike a conventional computer, DeepSouth hardware chips are designed to implement neural networks that model how synapses process information in the brain.

Computers of this type are called neuromorphic. They have been built before, but DeepSouth will be the largest, capable of performing 228 trillion synaptic operations per second. This figure is equal to the estimated number of synaptic operations in the human brain.

"For the first time, we will be able to simulate the activity of a neural network the size of a human brain in real time," explained Andre van Schaik, project leader at ICNS.

At the same time, DeepSouth will not be more powerful than existing supercomputers. Its main advantage is the ability to simulate the activity of the biological brain.

Supercomputers are so powerful that they have already become one of the largest energy consumers on the planet. At the same time, the extremely powerful human brain consumes the equivalent of energy slightly more than an ordinary light bulb.

According to scientists, this difference in energy consumption is partly due to the fact that traditional computers move information between the processor and memory in a fast sequence while processing data, while the neuromorphic architecture performs many operations in parallel with much less data movement.

In addition, neuromorphic neural networks work only when there is a change in the input data, while conventional computers work continuously in the background. This difference also affects how much energy the devices consume.

DeepSouth will be especially useful for those studying neuroscience or those who want to prototype new engineering solutions in the field of artificial intelligence.

In addition, if it paves the way for much greater energy efficiency in computing, the technology could later be adapted for use in drones and robots, making them more autonomous.

Subscribe to OBOZ.UA channels on Telegram and Viber to keep up with the latest events.