Charles Augustine at Intel's Circuit Research Laboratory in Hillsboro, Oregon, and a few of his colleagues unveil their design for a neuromorphic chip based on memristors and spin valves. Neurdons have heard this before... Once upon a time, in a galaxy far, far away....
... there was the DARPA SyNAPSE program. As many of you recall, several attempts at large-scale neuromorphic engineering have been made in the past. None met their goals. As such, SyNAPSE owes its existence to a number of recent game-changing developments. From HP Labs, the discovery of the memristor was one such keystone innovation. It took Greg Snider's 2007 work in Nanotechnology, however, to establish memristors as a viable platform for the implementation of self-organizing recurrent neural networks.
The new twist from Intel is the use of spin devices, or lateral spin valves. A spin valve is a device consisting of two or more conducting magnetic materials that align "up" or "down" depending on an external magnetic fields. Augustine and co argue that that the architecture they've designed works in a similar way to neurons and can therefore be used to test various ways of reproducing the brain's processing ability. Here is the link go the article. The claim of their work is huge energy saving: “We show that the spin-based neuromorphic designs can achieve 15X-300X lower computation energy,”.
This claim seems to me a bit out of the blue and unsupported, but it is surely an interesting paper, and something to read in the wake of interest started in 2007.