# Atomistic structure learning

## TL;DR

This paper introduces ASLA, an atomistic structure learning algorithm that uses reinforcement learning and neural networks to predict optimal atomic arrangements for desired properties without prior knowledge.

## Contribution

It presents a novel AI-based method for atomistic structure prediction that integrates first-principles quantum calculations with reinforcement learning.

## Key findings

- Successfully predicts grain boundaries in graphene
- Determines organic compound formations
- Identifies surface oxide structures

## Abstract

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.