Cross-platform frequency-domain physical neural networks with identical models and parameters
Abstract
Frequency-domain physical neural networks (PNNs) have emerged as a promising analog computing paradigm, achieving a reduced number of devices, enhanced robustness, and high inference accuracy. However, existing demonstrations across various physical domains, such as optics, acoustics, and electronics, are mostly specific to the target devices or platforms, usually requiring hardware-dependent models and post-fabrication parameter tuning. Here, we demonstrate cross-platform implementations of frequency-domain PNN across different wave-based computing platforms using identical, pre-trained parameters. By exploiting shared second-order nonlinear processes, the PNN model is implemented on optical, microwave, and acoustic-wave platforms without the need of fine tuning the parameters. Evaluated on a unified four-class classification task, the frequency-domain PNN achieves high inference accuracies of 97.6% in optics, 98.4% in electronics, and 98.2% in mechanics. Further, we systematically compare related performance metrics across these three distinct physical domains, paving the way to a universal neural network across different physical platforms.
Related
- Topological characterization of a reconfigurable synthetic-frequency SSH lattice on an integrated lithium-niobate platform
- Resonant multi-harmonic acousto-optics for programmable frequency control of visible light in a CMOS platform
- Design and Physical Constraints of Synthetic-Frequency Photonic Switching Fabrics
- Maximum likelihood, neural networks and posterior mean for inferring fractional Brownian motion parameters
- Inferring Nuclear Radial Parameters from Initial-State Information in Heavy-Ion Collisions Using Deep Neural Networks
BibTeX
@misc{xi2026crossplatform,
title = {{Cross-platform frequency-domain physical neural networks with identical models and parameters}},
author = {Zichen Xi and Jun Ji and Liyang Jin and Hsuan-Hao Lu and Wenjie Xiong and Linbo Shao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.01059},
eprint = {2610.01059},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.01059},
}
RIS
TY - UNPB
TI - Cross-platform frequency-domain physical neural networks with identical models and parameters
AU - Zichen Xi
AU - Jun Ji
AU - Liyang Jin
AU - Hsuan-Hao Lu
AU - Wenjie Xiong
AU - Linbo Shao
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.01059
UR - https://arxiv.org/abs/2610.01059
AB - Frequency-domain physical neural networks (PNNs) have emerged as a promising analog computing paradigm, achieving a reduced number of devices, enhanced robustness, and high inference accuracy. However, existing demonstrations across various physical domains, such as optics, acoustics, and electronics, are mostly specific to the target devices or platforms, usually requiring hardware-dependent models and post-fabrication parameter tuning. Here, we demonstrate cross-platform implementations of frequency-domain PNN across different wave-based computing platforms using identical, pre-trained parameters. By exploiting shared second-order nonlinear processes, the PNN model is implemented on optical, microwave, and acoustic-wave platforms without the need of fine tuning the parameters. Evaluated on a unified four-class classification task, the frequency-domain PNN achieves high inference accuracies of 97.6% in optics, 98.4% in electronics, and 98.2% in mechanics. Further, we systematically compare related performance metrics across these three distinct physical domains, paving the way to a universal neural network across different physical platforms.
ER -