NOMAD 2018 Kaggle Competition: Solving Materials Science Challenges Through Crowd Sourcing

TL;DR

This paper summarizes the top approaches from the NOMAD 2018 Kaggle competition, which aimed to identify the best machine learning models for predicting key properties of materials relevant to optoelectronic applications, using a dataset of 3,000 compounds.

Contribution

It introduces a new crystal graph representation for ML of materials properties and compares different models and representations through a comprehensive benchmarking study.

Findings

The crystal graph approach achieved top performance.

Combining multiple descriptors improved prediction accuracy.

Representation choice influences model performance more than regression method.

Abstract

Machine learning (ML) is increasingly used in the field of materials science, where statistical estimates of computed properties are employed to rapidly examine the chemical space for new compounds. However, a systematic comparison of several ML models for this domain has been hindered by the scarcity of appropriate datasets of materials properties, as well as the lack of thorough benchmarking studies. To address this, a public data-analytics competition was organized by the Novel Materials Discovery (NOMAD) Centre of Excellence and hosted by the on-line platform Kaggle using a dataset of $3\,000$ (Al$_x$ Ga$_y$ In$_z$)$_2$ O$_3$ compounds (with $x+y+z = 1$). The aim of this challenge was to identify the best ML model for the prediction of two key physical properties that are relevant for optoelectronic applications: the electronic band gap energy and the crystalline formation energy. In this contribution, we present a summary of the top three ML approaches of the competition including the 1st place solution based on a crystal graph representation that is new for ML of the properties of materials. The 2nd place model combined many candidate descriptors from a set of compositional, atomic environment-based, and average structural properties with the light gradient-boosting machine regression model. The 3rd place model employed the smooth overlap of atomic positions representation with a neural network. To gain insight into whether the representation or the regression model determines the overall model performance, nine ML models obtained by combining the representations and regression models of the top three approaches were compared by looking at the correlations among prediction errors. At fixed representation, the largest correlation is observed in predictions made with kernel ridge regression and neural network, reflecting a similar performance on the same test set samples.