Physics-Inspired Neural Networks for Efficient Device Compact Modeling

TLDR

High‑quality device compact models are essential for linking device science to applications, yet most rely on multilayer perceptron neural networks that often produce unphysical behavior, a problem that physics‑inspired neural networks aim to eliminate. The authors propose a physics‑inspired neural network (Pi‑NN) for compact modeling of electronic devices. Pi‑NN incorporates fundamental device physics into its architecture to prevent unphysical behavior and generate smooth, accurate models. Using Pi‑NN, smooth, accurate, and computationally efficient device models can be learned from discrete data points, highlighting a promising future for neural‑network‑based compact modeling.

Abstract

We present a novel physics-inspired neural network (Pi-NN) approach for compact modeling. Development of high-quality compact models for devices is a key to connect device science with applications. One recent approach is to treat compact modeling as a regression problem in machine learning. The most common learning algorithm to develop compact models is the multilayer perceptron (MLP) neural network. However, device compact models derived using the MLP neural networks often exhibit unphysical behavior, which is eliminated in the Pi-NN approach proposed in this paper, since the Pi-NN incorporates fundamental device physics. As a result, smooth, accurate, and computationally efficient device models can be learned from discrete data points by using Pi-NN. This paper sheds new light on the future of the neural network compact modeling.