The impending activation of a supercomputer in Australia next year puts into reality the once-fictional scenario of machines emulating human intelligence.
Called DeepSouth, this groundbreaking system is designed to simulate human brain synapses at full scale using spiking neural networks on its chips, in a bid to decipher how the brain processes vast amounts of information with minimal power consumption.
Developed jointly by researchers at the International Centre for Neuromorphic Systems (ICNS) at Western Sydney University, Intel, and Dell, DeepSouth’s neuromorphic hardware enables it to emulate large networks of spiking neurons at 228 trillion synaptic operations per second, rivaling the estimated rate of operations in the human brain.
“DeepSouth is purpose-built to operate like networks of neurons, requiring less power and enabling greater efficiencies,” said ICNS Director Professor Andre van Schaik. “This contrasts with supercomputers optimised for more traditional computing loads, which are power-hungry.”
Using the system’s spiking neural networks, which mimics real biological processes, DeepSouth paves the way for better understanding of the brain and the development of brain-scale computing applications in diverse fields, such as sensing, biomedical, robotics, space, and large-scale AI applications.
According to Professor Schaik, “our system will change that [simulating brain-like networks at scale]. This platform will progress our understanding of the brain.”