A domain knowledge enhanced machine learning method to predict the properties of halide double perovskite A₂B⁺B³⁺X₆

Abstract

Material datasets are high-dimensional and high-noise, which makes most machine learning (ML) methods inefficient. We present a new framework which embeds material domain knowledge into the ML method. By doing so, we illustrate its role and improve the prediction accuracy of 540 perovskite materials.