Neural-inspired method using time and time delay to perform tasks such as sorting

What is it about?

Here we provide a technique to compute simple, fundamental functions, which are part of larger more complicated algorithms, using artificial spiking neurons as the units of computation. This framework allows us to explore how these algorithms might be achieved using more biologically realistic elements and compare them to standard (non-neural-inspired) algorithms performing the same tasks.

Featured Image

Why is it important?

This work explores the crossroads between traditional digital and biologically-inspired computation, at a time when new neuromorphic computing devices are being developed without a clear path on how to make the best use of them (other than re-implementing digital logic using neuromorphic elements). Our paper is part of a body of research looking at it from the other direction, where we examine what can be achieved using these small, light and low-power devices.