MIT proton resistors enable deep learning to rise, in analog

A team of researchers from the Massachusetts Institute of Technology (MIT) is working on a new design of hardware resistance for the next era of scaling electronics — particularly in artificial intelligence processing tasks such as machine learning and neural networks.

However, in what may seem like a throwback (if there could even be a throwback to the future), their work focuses more on analog design than digital in nature. Enter programmable proton resistors — designed to speed up AI networks by simulating our neurons (and their connected synapses) as we speed them up a million times — and that’s the actual number, not just an exaggeration.

#MIT #proton #resistors #enable #deep #learning #rise #analog

Leave a Comment

Your email address will not be published.