...Good morning, CIOs. An experimental research system announced by Intel Wednesday has a computational capacity of about 100 million neurons, roughly equivalent to a mole-rat’s brain. And, yes, that’s a good thing if you are into artificial intelligence.
The system, called Pohoiki Springs and available later this month to a host of researchers, government labs and about a dozen companies, represents a new approach to working with neural networks called neuromorphic computing.
Why it’s special. Mike Davies, director of Intel’s Neuromorphic Computing Lab, tells CIO Journal’s Sara Castellanos that neuromorphic computing makes it possible to train machine-learning models using a fraction of the data it takes to train them on traditional computing hardware. Models learn similarly to the way human babies learn, by seeing an image or object once and being able to recognize it forever.
No comments:
Post a Comment