The first supercomputer that imitates the capabilities of the human brain about to see the light
A team of researchers from Western Sydney University, Australia, is on the verge of achieving a revolutionary milestone in the field of artificial intelligence and neuroscience. This is DeepSouth, the First neuromorphic supercomputer designed to mimic human brain processing capabilities.
Its name, in addition to being derived from its location, is a nod to both IBM’s TrueNorth system, which led efforts to build machines capable of imitating large networks of neurons, and Deep Blue, the world’s first computer chess champion.
What makes it unique? Your ability to perform 228 trillion operations per second, equaling the efficiency of the brain, which, with a modest 20 watts of powerexecutes processes equivalent to an exaflop.
The arrival of this beast, scheduled for next year, promises to open new possibilities for understanding the efficiency of brains in processing information with a minimum amount of energy. They not only want to make a more powerful machine, but also discover the secrets that could change the way we understand the brain and artificial intelligence.
How will DeepSouth work and what impact will it have on scientific research?
However, André van Schaik, a professor at Western Sydney University and director of its International Center for Neuromorphic Systems, says efforts are currently underway to understand what makes the brain so extraordinary and, more specifically, what facilitates its impressive capabilities using Neurons advance slowly.
This is mainly because you cannot simulate networks that function like the human brain at scale. The Australian team precisely aims to change that with DeepSouth.
Stands out for using a different approach: imitating biological processes instead of relying on conventional graphics processing units (GPUs) and central processing units (CPUs), thus paving the way toward the future creation of synthetic brains that can surpass current capabilities.
Of course, this novelty will allow further progress in the field of artificial intelligence. “This platform will improve our understanding of the brain and develop brain-scale computing applications in various fields, including sensing, biomedical, robotics, space and large-scale artificial intelligence applications,” they explain.
