A Neural Network Based on Synchronized Pairs of Nano-Oscillators

TL;DR

This paper proposes a novel neural network architecture using synchronized nano-oscillators that can implement logic gates like XOR and achieve competitive classification results, offering a promising alternative for efficient neural computation.

Contribution

It introduces a new oscillator-based neuron model leveraging synchronization dynamics, capable of implementing complex logic and performing classification tasks.

Findings

The oscillator-based neuron can implement logic gates including XOR.

Simulated neural networks achieve standard classification performance.

Performance remains robust under oscillator phase noise.

Abstract

Artificial neural networks are intensively used to perform cognitive tasks such as image classification on traditional computers. With the end of CMOS scaling and increasing demand for efficient neural networks, alternative architectures implementing neural functions efficiently are being studied. This study leverages the demonstrated frequency tuning capabilities of compact nano-oscillators and their synchronization dynamics to implement a neuron using a pair of synchronized oscillators, and which features an unconventional response curve. We show that this compact neuron can naturally implement generic logic gates, including XOR. A simulated oscillator-based neural network is then shown to achieve results equivalent to standard approaches on two reference classification tasks. Finally, the performance of the system is evaluated in the presence of oscillator phase noise, an important issue of oscillating nanodevices. These results open the way for the design of alternative architectures adapted to efficient neural network execution.