Digital-analog concept for superconducting perceptron-like neural networks
Abstract
A promising route to superconducting artificial neural networks is a hybrid digital-Analog architecture that combines digital single-flux-quantum (SFQ) communication with compact Analog nonlinear processing. The study focused on the dynamic conversion of a discrete signal passing through a digital-to-Analog-to-digital (DAD) converter, in which the role of the Analog cell was performed by a Σ-neuron with a nonlinear transfer function -- the basic cell of perceptron-like neural networks. Furthermore, the DAD converter, the elementary functional block of the hybrid architecture, combines a digital-to-analog converter (DAC) and an Analog-to-digital converter (ADC), and re-encodes the Analog Σ-neuron waveforms as an SFQ pulse sequence. Circuit-level simulations demonstrate how input values encoded by SFQ pulse trains are converted into analog signal levels, transformed by the Σ-neuron, and mapped back to pulse-based outputs. As a key experimental step, we fabricated and characterised a redesigned Σ-neuron and measured a sigmoid-like transfer characteristic suitable for activation-function implementation. The extracted response was incorporated into system-level simulations to assess the influence of realistic device parameters on the conversion process. We delineate the operating-range matching requirements for the DAC, neuron, and ADC blocks, supporting the feasibility of the proposed interface as a building block for perceptron-like superconducting neural networks with digital inputs and outputs. Finally, we developed two perceptron networks, one using a mathematical sigmoid activation and the other the measured Σ-neuron transfer characteristic, which reached classification accuracies of 97.0% and 91.9%, respectively, on the MNIST handwritten digit dataset.
Versions
Related
- Analog quantum simulation of ϕ4 field theory with a superconducting transmission line
- Early Fault-Tolerant Magic State Synthesis and Spacetime Efficient Analog Rotations with Color Code on a Superconducting Quantum Processor
- Nonlinear pluggable optics: Digital signal processing-free Intensity Modulated Direct Detection links using analog photonic Next Generation Reservoir Computing
- Integrated Non-Sliced Optical Arbitrary-Waveform Measurement (OAWM) Engine Enabling 320 GHz Photonic-Electronic Analog-to-Digital Conversion
- Analog-to-digital conversion in the quantum regime
BibTeX
@misc{schegolev2026digitalanalog,
title = {{Digital-analog concept for superconducting perceptron-like neural networks}},
author = {Andrey E. Schegolev and Vsevolod I. Ruzhickiy and Georgy I. Gubochkin and Alexander S. Ionin and Ivan A. Nazhestkin and Mikhail Y. Fominskii and Lyudmila V. Filippenko and Igor I. Soloviev and Maxim V. Tereshonok and Nikolay V. Klenov},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21613},
eprint = {2609.21613},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21613},
}
RIS
TY - UNPB
TI - Digital-analog concept for superconducting perceptron-like neural networks
AU - Andrey E. Schegolev
AU - Vsevolod I. Ruzhickiy
AU - Georgy I. Gubochkin
AU - Alexander S. Ionin
AU - Ivan A. Nazhestkin
AU - Mikhail Y. Fominskii
AU - Lyudmila V. Filippenko
AU - Igor I. Soloviev
AU - Maxim V. Tereshonok
AU - Nikolay V. Klenov
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21613
UR - https://arxiv.org/abs/2609.21613
AB - A promising route to superconducting artificial neural networks is a hybrid digital-Analog architecture that combines digital single-flux-quantum (SFQ) communication with compact Analog nonlinear processing. The study focused on the dynamic conversion of a discrete signal passing through a digital-to-Analog-to-digital (DAD) converter, in which the role of the Analog cell was performed by a Σ-neuron with a nonlinear transfer function -- the basic cell of perceptron-like neural networks. Furthermore, the DAD converter, the elementary functional block of the hybrid architecture, combines a digital-to-analog converter (DAC) and an Analog-to-digital converter (ADC), and re-encodes the Analog Σ-neuron waveforms as an SFQ pulse sequence. Circuit-level simulations demonstrate how input values encoded by SFQ pulse trains are converted into analog signal levels, transformed by the Σ-neuron, and mapped back to pulse-based outputs. As a key experimental step, we fabricated and characterised a redesigned Σ-neuron and measured a sigmoid-like transfer characteristic suitable for activation-function implementation. The extracted response was incorporated into system-level simulations to assess the influence of realistic device parameters on the conversion process. We delineate the operating-range matching requirements for the DAC, neuron, and ADC blocks, supporting the feasibility of the proposed interface as a building block for perceptron-like superconducting neural networks with digital inputs and outputs. Finally, we developed two perceptron networks, one using a mathematical sigmoid activation and the other the measured Σ-neuron transfer characteristic, which reached classification accuracies of 97.0% and 91.9%, respectively, on the MNIST handwritten digit dataset.
ER -