A neuromorphic network for generic multivariate data classification

TLDR

Computational neuroscience seeks to uncover fundamental principles of brain computations. The study applies biologically inspired principles to develop a brain‑like processing system. The authors use neuromorphic microchip hardware to build a neural network modeled on insect sensory processing. The neuromorphic network successfully classifies generic multidimensional data, proving that analog microcircuits can perform real‑world computing tasks while highlighting benefits and challenges.

Abstract

Significance One primary goal of computational neuroscience is to uncover fundamental principles of computations that are performed by the brain. In our work, we took direct inspiration from biology for a technical application of brain-like processing. We make use of neuromorphic hardware—electronic versions of neurons and synapses on a microchip—to implement a neural network inspired by the sensory processing architecture of the nervous system of insects. We demonstrate that this neuromorphic network achieves classification of generic multidimensional data—a widespread problem with many technical applications. Our work provides a proof of concept for using analog electronic microcircuits mimicking neurons to perform real-world computing tasks, and it describes the benefits and challenges of the neuromorphic approach.