Deep Learning to Automate Parameter Extraction and Model Fitting of Two-Dimensional Transistors

TL;DR

This paper introduces a deep learning method for automated extraction of physical parameters from 2D transistor measurements, significantly reducing data requirements and enabling accurate, scalable reverse-engineering of device characteristics.

Contribution

The authors develop a deep learning framework that leverages physics-based simulations and data augmentation to efficiently extract multiple device parameters from electrical data.

Findings

Achieved median R^2 of 0.99 on experimental WS2 transistors.

Reduced training data needs by over 40 times compared to previous methods.

Successfully reverse-engineered 35 parameters in high-electron-mobility transistors.

Abstract

We present a deep learning approach to extract physical parameters (e.g., mobility, Schottky contact barrier height, defect profiles) of two-dimensional (2D) transistors from electrical measurements, enabling automated parameter extraction and technology computer-aided design (TCAD) fitting. To facilitate this task, we implement a simple data augmentation and pre-training approach by training a secondary neural network to approximate a physics-based device simulator. This method enables high-quality fits after training the neural network on electrical data generated from physics-based simulations of ~500 devices, a factor >40$\times$ fewer than other recent efforts. Consequently, fitting can be achieved by training on physically rigorous TCAD models, including complex geometry, self-consistent transport, and electrostatic effects, and is not limited to computationally inexpensive compact models. We apply our approach to reverse-engineer key parameters from experimental monolayer WS$_2$ transistors, achieving a median coefficient of determination ($R^2$) = 0.99 when fitting measured electrical data. We also demonstrate that this approach generalizes and scales well by reverse-engineering electrical data on high-electron-mobility transistors while fitting 35 parameters simultaneously. To facilitate future research on deep learning approaches for inverse transistor design, we have published our code and sample data sets online.