Solving a steady-state PDE using spiking networks and neuromorphic hardware

TL;DR

This paper demonstrates solving a steady-state heat equation using spiking neural networks on neuromorphic hardware, showcasing the potential for physics-based computations and proposing a scalable benchmark for such systems.

Contribution

It introduces a novel method to solve PDEs with spiking networks and implements it on IBM TrueNorth and Intel Loihi hardware, expanding neuromorphic applications.

Findings

Successful implementation on TrueNorth and Loihi hardware

Effective use of stochastic neuron behavior for PDE solving

Proposes a scalable benchmark for neuromorphic systems

Abstract

The widely parallel, spiking neural networks of neuromorphic processors can enable computationally powerful formulations. While recent interest has focused on primarily machine learning tasks, the space of appropriate applications is wide and continually expanding. Here, we leverage the parallel and event-driven structure to solve a steady state heat equation using a random walk method. The random walk can be executed fully within a spiking neural network using stochastic neuron behavior, and we provide results from both IBM TrueNorth and Intel Loihi implementations. Additionally, we position this algorithm as a potential scalable benchmark for neuromorphic systems.