Instant prediction of relaxation in moiré superlattices using neural networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Belonovskii, Aleksei V., Girshova, Elizaveta I., Lähderanta, Erkki, Kaliteevski, Mikhail
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908542739415040
author Belonovskii, Aleksei V.
Girshova, Elizaveta I.
Lähderanta, Erkki
Kaliteevski, Mikhail
author_facet Belonovskii, Aleksei V.
Girshova, Elizaveta I.
Lähderanta, Erkki
Kaliteevski, Mikhail
contents The relaxation of moiré superlattices in twisted bilayers of transition metal dichalcogenides (TMDs) has been modeled using a set of neural-network-based approaches. We implemented and compared several architectures, including (i) an interpolator combined with an autoencoder, (ii) an interpolator combined with a decoder, (iii) a direct generator mapping input parameters to displacement fields, and (iv) a physics-informed neural network (PINN). Among these, the direct generator architecture demonstrated the best performance, achieving machine-level precision with minimal training data. Remarkably, once trained, this simple fully connected network is able to predict the full displacement field of a moiré bilayer within a fraction of a second, whereas conventional continuum simulations require hours or even days. This finding highlights the low-dimensional nature of the relaxation process and establishes neural networks as a practical and efficient alternative to ab initio approaches for rapid modeling and high-throughput screening of 2D twisted heterostructures.
format Preprint
id arxiv_https___arxiv_org_abs_2509_13147
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Instant prediction of relaxation in moiré superlattices using neural networks
Belonovskii, Aleksei V.
Girshova, Elizaveta I.
Lähderanta, Erkki
Kaliteevski, Mikhail
Disordered Systems and Neural Networks
The relaxation of moiré superlattices in twisted bilayers of transition metal dichalcogenides (TMDs) has been modeled using a set of neural-network-based approaches. We implemented and compared several architectures, including (i) an interpolator combined with an autoencoder, (ii) an interpolator combined with a decoder, (iii) a direct generator mapping input parameters to displacement fields, and (iv) a physics-informed neural network (PINN). Among these, the direct generator architecture demonstrated the best performance, achieving machine-level precision with minimal training data. Remarkably, once trained, this simple fully connected network is able to predict the full displacement field of a moiré bilayer within a fraction of a second, whereas conventional continuum simulations require hours or even days. This finding highlights the low-dimensional nature of the relaxation process and establishes neural networks as a practical and efficient alternative to ab initio approaches for rapid modeling and high-throughput screening of 2D twisted heterostructures.
title Instant prediction of relaxation in moiré superlattices using neural networks
topic Disordered Systems and Neural Networks
url https://arxiv.org/abs/2509.13147