The document discusses recent developments in materials design using deep learning (DL) and quantum computation. It introduces several DL and quantum computation models developed at NIST including:
- ALIGNN, a graph neural network model that predicts materials properties from crystal structure.
- AtomVision, a DL framework for analyzing scanning probe microscopy and electron microscopy images of materials.
- AtomQC, which uses variational quantum algorithms like VQE to perform quantum simulations of materials on quantum computers.
The models have been applied to datasets containing thousands of materials to predict properties and analyze experimental images. Future work aims to integrate DL and quantum computation for accelerated materials discovery and design.