# Towards the computational experiment

## TL;DR

This paper discusses current limitations in computational material simulations and explores how machine learning interatomic potentials are transforming the field and shaping its future development.

## Contribution

It provides an overview of the field's challenges and highlights the impact of machine learning potentials on future research directions.

## Key findings

- Machine learning potentials are improving simulation accuracy.
- Emerging methods are expanding the scope of computational experiments.
- The field faces limitations that machine learning aims to address.

## Abstract

We give a brief account of the current limitations and possibilities in the field of computational simulation of materials. We then focus on the effect that the emergence of machine learning interatomic potentials is having on the field and how it will affect its evolution in the near future.