Elf autoencoder for unsupervised exploration of flat-band materials using electronic band structure fingerprints

Artem Mishchenko, Henry Kelbrick Pentz, Thomas Warford, Ivan Timokhin, Hongpeng Zhou, Qian Yang, Anupam Bhattacharya

Keywords: AI, machine learning, autoencoder, convolutional neural network, electronic band structure, flat-band materials, unsupervised learning, clustering, density functional theory, materials discovery, 2D materials, band topology, high-throughput screening, feature extraction, deep learning

TL;DR: We present elf, an AI-powered autoencoder that extracts electronic band structure fingerprints to autonomously discover and classify flat-band materials, revealing hidden chemical trends beyond traditional methods.

Confirmation Of Submission Requirements: I submit a previously published paper. It was published in an archival peer–reviewed venue on or after September 8th 2024, I specify the DOI in the field below, and I submit the camera-ready version of the paper.

DOI: https://doi.org/10.1038/s42005-025-01936-2

Submission Number: 81