Physics-Informed Neural Networks for Device and Circuit Modeling: A Case Study of NeuroSPICE

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tung, Chien-Ting, Hu, Chenming
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913063685324800
author Tung, Chien-Ting
Hu, Chenming
author_facet Tung, Chien-Ting
Hu, Chenming
contents We present NeuroSPICE, a physics-informed neural network (PINN) framework for device and circuit simulation. Unlike conventional SPICE, which relies on time-discretized numerical solvers, NeuroSPICE leverages PINNs to solve circuit differential-algebraic equations (DAEs) by minimizing the residual of the equations through backpropagation. It models device and circuit waveforms using analytical equations in time domain with exact temporal derivatives. While PINNs do not outperform SPICE in speed or accuracy during training, they offer unique advantages such as surrogate models for design optimization and inverse problems. NeuroSPICE's flexibility enables the simulation of emerging devices, including highly nonlinear systems such as ferroelectric memories.
format Preprint
id arxiv_https___arxiv_org_abs_2512_23624
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Physics-Informed Neural Networks for Device and Circuit Modeling: A Case Study of NeuroSPICE
Tung, Chien-Ting
Hu, Chenming
Artificial Intelligence
Applied Physics
We present NeuroSPICE, a physics-informed neural network (PINN) framework for device and circuit simulation. Unlike conventional SPICE, which relies on time-discretized numerical solvers, NeuroSPICE leverages PINNs to solve circuit differential-algebraic equations (DAEs) by minimizing the residual of the equations through backpropagation. It models device and circuit waveforms using analytical equations in time domain with exact temporal derivatives. While PINNs do not outperform SPICE in speed or accuracy during training, they offer unique advantages such as surrogate models for design optimization and inverse problems. NeuroSPICE's flexibility enables the simulation of emerging devices, including highly nonlinear systems such as ferroelectric memories.
title Physics-Informed Neural Networks for Device and Circuit Modeling: A Case Study of NeuroSPICE
topic Artificial Intelligence
Applied Physics
url https://arxiv.org/abs/2512.23624