Learning atoms for materials discovery

By Quan Zhou, Peizhe Tang, Shenxiu Liu, Jinbo Pan, Q. Yan, Shoucheng Zhang. 2018

195 citations recorded by the source corpus.

Significance Motivated by the recent achievements of artificial intelligence (AI) in linguistics, we design AI to learn properties of atoms from materials data on its own. Our work realizes knowledge representation of atoms via computers and could serve as a foundational step toward materials discovery and design fully based on machine learning. Exciting advances have been made in artificial intelligence (AI) during recent decades. Among them, applications of machine learning (ML) and deep learning techniques brought human-competitive performances in various tasks of fields, including image recognition, speech recognition, and natural language understanding. Even in Go, the ancient game of profound complexity, the AI player has already beat human world champions convincingly with and without learning from the human. In this work, we show that our unsupervised machines (Atom2Vec) can learn the basic properties of atoms by themselves from the extensive database of known compounds and materials. These learned properties are represented in terms of high-dimensional vectors, and clustering of atoms in vector space classifies them into meaningful groups consistent with human knowledge. We use the atom vectors as basic input units for neural networks and other ML models designed and trained to predict materials properties, which demonstrate significant accuracy.

Subfields: Biomedical Engineering, Materials Chemistry, AI for Physical Sciences

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Jure Leskovec
2. Percy Liang
3. Vijay S. Pande
4. Xianglin Liu
5. Zongrui Pei
6. Jiaxin Zhang

In Materials Chemistry

1. Jure Leskovec
2. Chi Chen
3. Shyue Ping Ong
4. Marinka Zitnik
5. Bowen Liu
6. Weihua Hu
7. Joseph Gomes
8. Percy Liang

In AI for Physical Sciences

1. Yunxing Zuo
2. Chi Chen
3. Percy Liang
4. Xianglin Liu
5. Shyue Ping Ong
6. Zongrui Pei
7. Jiaxin Zhang

Citations in the corpus

• Strategies for Pre-training Graph Neural Networks (2019, background): “A number of recent works have also explored how node embeddings generalize across tasks (Jaeger et al., 2018; Zhou et al., 2018; Chakravarti, 2018; Narayanan et al., 2016).”
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals (2018, methodology): “55 To this end, we have extracted the elemental embeddings from the MEGNet model for crystal formation energy.”
• Pre-training Graph Neural Networks (2019, background): “It should be noted that similar intuition has been explored in the literature [32, 19, 63].”
• Machine Learning for High-entropy Alloys: Progress, Challenges and Opportunities (2022, background): “As a categorical variable, the atomic species can be represented with one-hot encoding, element embedding [143, 70], or simply assigning a unique weight for each atomic species.”