Connectivity Optimized Nested Graph Networks for Crystal Structures

Ruff, Robin; Reiser, Patrick; Stühmer, Jan; Friederich, Pascal

doi:10.48550/arxiv.2302.14102

Cited by 4 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these particular tasks do not have a structure provided and only chemical composition is known. Meanwhile, for tasks with >10,000 samples, the (related) coNGN 58 and coGN 58 algorithms lead 5 out of 6 tasks. This separation based on sample size (with different leading algorithms at that time) was observed in the original matbench paper 53 .…”

Section: Steady Gains In Accuracymentioning

confidence: 99%

Machine learning in materials research: developments over the last decade and challenges for the future

Jain

2024

Preprint

View full text Add to dashboard Cite

The number of studies that apply machine learning (ML) to materials science has been growing at a rate of approximately 1.67 times per year over the past decade. In this review, I examine this growth in various contexts. First, I present an analysis of the most commonly used tools (software, databases, materials science methods, and ML methods) used within papers that apply ML to materials science. The analysis demonstrates that despite the growth of deep learning techniques, the use of classical machine learning is still dominant as a whole. It also demonstrates how new research can effectively build upon past research, particular in the domain of ML models trained on density functional theory calculation data. Next, I present the progression of best scores as a function of time on the matbench materials science benchmark for formation enthalpy prediction. In particular, a dramatic improvement of 7 times reduction in error is obtained when progressing from feature-based methods that use conventional ML (random forest, support vector regression, etc.) to the use of graph neural network techniques. Finally, I provide views on future challenges and opportunities, focusing on data size and complexity, extrapolation, interpretation, access, and relevance.

show abstract

Section: Steady Gains In Accuracymentioning

confidence: 99%

Machine learning in materials research: developments over the last decade and challenges for the future

Jain

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Atoms sharing a face are considered neighbors, and edges are consequently drawn between them (see the Appendix for illustrative figures and details). Nevertheless, the Voronoi-based method still has drawbacks: (1) the number of edges and their respective pairwise distances can be highly variable and may exhibit significant fluctuations across atoms (Ruff et al 2023); (2) computational efficiency diminishes noticeably in spaces beyond 2D (Vassiliades, Chatzilygeroudis, and Mouret 2017).…”

Section: Limitations Of Existing Graph Constructionsmentioning

confidence: 99%

Conformal Crystal Graph Transformer with Robust Encoding of Periodic Invariance

Wang,

Kong,

Gregoire

et al. 2024

AAAI

View full text Add to dashboard Cite

Machine learning techniques, especially in the realm of materials design, hold immense promise in predicting the properties of crystal materials and aiding in the discovery of novel crystals with desirable traits. However, crystals possess unique geometric constraints—namely, E(3) invariance for primitive cell and periodic invariance—which need to be accurately reflected in crystal representations. Though past research has explored various construction techniques to preserve periodic invariance in crystal representations, their robustness remains inadequate. Furthermore, effectively capturing angular information within 3D crystal structures continues to pose a significant challenge for graph-based approaches. This study introduces novel solutions to these challenges. We first present a graph construction method that robustly encodes periodic invariance and a strategy to capture angular information in neural networks without compromising efficiency. We further introduce CrystalFormer, a pioneering graph transformer architecture that emphasizes angle preservation and enhances long-range information. Through comprehensive evaluation, we verify our model's superior performance in 5 crystal prediction tasks, reaffirming the efficiency of our proposed methods.

show abstract

“…In the realm of SPMs, numerous studies have been conducted to fully exploit structure− property knowledge using different model structures or feature representations. This encompasses efforts like CGCNN, 15 SchNet, 16 ALIGNN, 13 and coGN 17 which have progressively minimized the error in predicting material properties. At the same time, in the field of CPMs, there is an abundance of commendable work steadily reducing property prediction errors.…”

Section: ■ Introductionmentioning

confidence: 99%

Addressing the Accuracy-Cost Trade-off in Material Property Prediction Using a Teacher-Student Strategy

Zhu,

Xin,

Zheng

et al. 2024

J. Chem. Theory Comput.

View full text Add to dashboard Cite

Deep learning has catalyzed a transformative shift in material discovery, offering a key advantage over traditional experimental and theoretical methods by significantly reducing associated costs. Models adept at predicting properties from chemical compositions alone do not require structural information. However, this cost-efficient approach compromises model precision, particularly in Chemical Composition-based Property Prediction Models (CPMs), which are notably less accurate than Structure-based Property Prediction Models (SPMs). Addressing this challenge, our study introduces a novel Teacher-Student (TS) strategy, where a pretrained SPM serves as an instructive 'teacher' to enhance the CPM's precision. This TS strategy successfully harmonizes low-cost exploration with high accuracy, achieving a significant 47.1% reduction in relative error in scenarios involving 100 data entries. We also evaluate the effectiveness of the proposed strategy by employing perovskites as a case study. This method represents a significant advancement in the exploration and identification of valuable materials, leveraging CPM's potential while overcoming its precision limitations.

show abstract

Connectivity Optimized Nested Graph Networks for Crystal Structures

Cited by 4 publications

References 25 publications

Machine learning in materials research: developments over the last decade and challenges for the future

Machine learning in materials research: developments over the last decade and challenges for the future

Conformal Crystal Graph Transformer with Robust Encoding of Periodic Invariance

Addressing the Accuracy-Cost Trade-off in Material Property Prediction Using a Teacher-Student Strategy

Contact Info

Product

Resources

About