Predicting Superconductivity via Machine Learning Methods

Superconducting materials present the rare opportunity to revolutionize virtually every aspect of human life, including improved transportation, more efficient energy transfer, and faster computers. With the help of Artificial Intelligence (AI) and Machine Learning (ML) models, we can study what properties influence the superconductivity of a compound and predict what experimental and theoretical compounds will perform as efficient superconductors. Our control substance is the popular YBCO compound. Using experimental data gathered in Adelphi's first comprehensive materials science endeavor, with the aid of encoded, theoretical deep learning models, our project seeks to investigate the Meissner Effect. Our study tackles the question of what makes a superconductor a superconductor, with the ultimate goal of discovering additional high temperature superconductors to be synthesized in the lab.