A multi-category inverse design neural network and its application to diblock copolymers

TL;DR

This paper introduces a multi-category inverse design neural network that accurately predicts physical parameters of periodic structures, specifically applied to diblock copolymers, with an innovative data augmentation ensuring invariance.

Contribution

The work presents a novel multi-category neural network architecture combined with a reciprocal-space data augmentation method for inverse design of periodic structures.

Findings

High accuracy in predicting physical parameters for desired structures

Effective rotation and translation invariance achieved through data augmentation

Method applicable to other inverse design problems

Abstract

In this work, we design a multi-category inverse design neural network to map ordered periodic structure to physical parameters. The neural network model consists of two parts, a classifier and Structure-Parameter-Mapping (SPM) subnets. The classifier is used to identify structure, and the SPM subnets are used to predict physical parameters for desired structures. We also present an extensible reciprocal-space data augmentation method to guarantee the rotation and translation invariant of periodic structures. We apply the proposed network model and data augmentation method to two-dimensional diblock copolymers based on the Landau-Brazovskii model. Results show that the multi-category inverse design neural network is high accuracy in predicting physical parameters for desired structures. Moreover, the idea of multi-categorization can also be extended to other inverse design problems.