A Neuromorphic Implementation of the DBSCAN Algorithm

TL;DR

This paper presents two neuromorphic implementations of the DBSCAN clustering algorithm using spiking neural networks, enabling fast and efficient clustering with pipelining and size trade-offs.

Contribution

It introduces two novel neuromorphic constructions for DBSCAN, one flat and fast, the other systolic and compact, with detailed specifications and open-source code.

Findings

Flat construction computes in five timesteps

Pipelining allows new calculations every timestep

Systolic construction uses smaller networks with multi-timestep inputs

Abstract

DBSCAN is an algorithm that performs clustering in the presence of noise. In this paper, we provide two constructions that allow DBSCAN to be implemented neuromorphically, using spiking neural networks. The first construction is termed "flat," resulting in large spiking neural networks that compute the algorithm quickly, in five timesteps. Moreover, the networks allow pipelining, so that a new DBSCAN calculation may be performed every timestep. The second construction is termed "systolic", and generates much smaller networks, but requires the inputs to be spiked in over several timesteps, column by column. We provide precise specifications of the constructions and analyze them in practical neuromorphic computing settings. We also provide an open-source implementation.