Atomistic structure learning
This is a foundational step toward direct manipulation of atoms for materials and drug design, potentially impacting physical chemistry and AI-driven discovery.
The authors tackled the problem of designing materials and drugs by introducing an atomistic structure learning algorithm (ASLA) that builds 2D compounds and layered structures atom by atom using a convolutional neural network and reinforcement learning, without prior data, and demonstrated it on diverse problems like grain boundaries in graphene and organic compound formation.
One endeavour of modern physical chemistry is to use bottom-up approaches to design materials and drugs with desired properties. Here we introduce an atomistic structure learning algorithm (ASLA) that utilizes a convolutional neural network to build 2D compounds and layered structures atom by atom. The algorithm takes no prior data or knowledge on atomic interactions but inquires a first-principles quantum mechanical program for physical properties. Using reinforcement learning, the algorithm accumulates knowledge of chemical compound space for a given number and type of atoms and stores this in the neural network, ultimately learning the blueprint for the optimal structural arrangement of the atoms for a given target property. ASLA is demonstrated to work on diverse problems, including grain boundaries in graphene sheets, organic compound formation and a surface oxide structure. This approach to structure prediction is a first step toward direct manipulation of atoms with artificially intelligent first principles computer codes.