Advanced memristors are a pathway to affordable human scale neuromorphic artificial general intelligence

It may be possible to create neuromorphic human-level Artificial General Intelligence within 5 to 15 years for 30 to 100 thousand dollars, of marginal cost using memristors.

The extreme complexity of the human cerebral cortex,featuring in particular ~10^14 synapses, makes the hardware implementation of neuromorphic networks with a comparable number of devices exceptionally challenging. One of the most prospective candidates to provide comparable complexity, while operating muchfaster and with manageable power dissipation, are so-called CrossNets based on hybrid CMOS/memristor circuits. In these circuits, the usual complementary metal-oxide-semiconductor (CMOS) stack is augmented with one or several crossbar layers, with adjustable two-terminal resistive devices (“memristors”) at each crosspoint. Recently, there was a significant progress in improvement of technology of fabrication of such memristive crossbars and their integration with CMOS circuits,including first demonstrations of their vertical integration. Separately, there have been several demonstrations of discrete memristors as artificial synapses for neuromorphic networks.Very recently such experiments were extended to crossbar arrays of phase-change memristive devices

Arxiv – Training and Operation of an Integrated Neuromorphic Network Based on Metal-Oxide Memristors by Prezioso, Merrikh-Bayat, Hoskins, Adam, Likharev, and Strukov:

“… a CrossNets based on a hybrid CMOS/memristor circuit with 5 layers of 30-nm-pitch crossbars, 2 memristors per synapse, and 10^4 synapses per neural cell would have an areal density of ~25 million cells per cm^2, i.e. higher than that in the human cerebral cortex, at comparable average connectivity. Estimates show that at the same time, such CrossNets may provide comparable power efficiency, at a much higher operation speed – for example, an intercell signal transfer delay of ~0.02 ms (cf. ~10 ms in biology) [my comment: i.e., 500 times faster] at a readily manageable energy dissipation rate of ~1 W/cm2.”

HP’s CTO showed off a wafer of memristor memory for getting to 100 TB drives. Neuromorphic human scale AFI would need memristors that mimicked synapses. Logic and memory mixed together with analog behavior