Autonomous Generation of Single Photon Emitting Materials

Abstract

The utilization of machine learning in Materials Science has highlighted that trained models' effectiveness is dependent on the quality and quantity of data utilized for training. Unlike fields such as image processing and natural language processing, there is limited availability of atomistic datasets, leading to biases in training datasets. Particularly in the domain of materials discovery, there exists an issue of continuity in atomistic datasets. Experimental data sourced from literature and patents is usually only available for a select number of atomistic data, resulting in bias in the training dataset. This study focuses on developing a language-based model for generating a synthetic dataset of quantum materials using a variational autoencoder approach. The study centers on generating a synthetic dataset of quantum materials specifically for quantum sensing applications, with a focus on two-level quantum molecules demonstrating dipole blockade. The proposed technique offers an improved sampling algorithm by incorporating newly generated materials into the sampling algorithm to create a more normally distributed dataset. Through this technique, the study was able to generate over 1,000,000 candidate quantum materials from a small dataset of only 3,000 materials. The generated dataset identified several iodine-containing molecules as promising single photon emitting materials for potential quantum sensing applications.