Tuesday, January 21, 2020

Beyond Moore's Law: Neuromorphic computing?

With the conceivable exhaustion of Moore’s Law – that the number of transistors on a microchip doubles every two years – the search is on for new paths that lead to reliable incremental processing gains over time.

One possibility is that machines inspired by how the brain works could take over, fundamentally shifting computing to a revolutionary new tier, according to an explainer study released this month by Applied Physics Reviews.

“Today’s state-of-the-art computers process roughly as many instructions per second as an insect brain,” say the paper’s authors Jack Kendall, of Rain Neuromorphics, and Suhas Kumar, of Hewlett Packard Labs. The two write that processor architecture must now be completely re-thought if Moore’s law is to be perpetuated, and that replicating the “natural processing system of a [human] brain” is the way forward.

To read this article in full, please click here


Thanks to Patrick Nelson (see source)

No comments:

Post a Comment