Superconducting Optoelectronic Neurons V: Networks and Scaling

TL;DR

This paper explores the design, physical scaling, and power efficiency of large-scale superconducting optoelectronic neural networks, demonstrating their potential for massive, high-speed, low-power neural systems.

Contribution

It introduces a scalable network construction method with fractal properties and analyzes physical size and power consumption for networks of various scales.

Findings

Networks with millions of neurons fit on a 300mm wafer.

Large networks can operate coherently at 1 MHz.

Power density remains low enough for liquid helium cooling.

Abstract

Networks of superconducting optoelectronic neurons are investigated for their near-term technological potential and long-term physical limitations. Networks with short average path length, high clustering coefficient, and power-law degree distribution are designed using a growth model that assigns connections between new and existing nodes based on spatial distance as well as degree of existing nodes. The network construction algorithm is scalable to arbitrary levels of network hierarchy and achieves systems with fractal spatial properties and efficient wiring. By modeling the physical size of superconducting optoelectronic neurons, we calculate the area of these networks. A system with 8100 neurons and 330,430 total synapses will fit on a 1\,cm $\times$ 1\,cm die. Systems of millions of neurons with hundreds of millions of synapses will fit on a 300\,mm wafer. For multi-wafer assemblies, communication at light speed enables a neuronal pool the size of a large data center comprising 100 trillion neurons with coherent oscillations at 1\,MHz. Assuming a power law frequency distribution, as is necessary for self-organized criticality, we calculate the power consumption of the networks. We find the use of single photons for communication and superconducting circuits for computation leads to power density low enough to be cooled by liquid $^4$He for networks of any scale.