Machine learning material properties from the periodic table using convolutional neural networks

Zheng, Xiaolong; Zheng, Peng; Zhang, Ruizhi

doi:10.1039/c8sc02648c

Cited by 90 publications

(53 citation statements)

References 34 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, like the previous work using the SuperCon database, the crystal structure information was not included, making it difficult to draw conclusions even the model error was low. In addition, similar approaches to projecting the periodic table to an image have been adopted by Zheng et al to find novel X 2 YZ compounds. The prediction of T cr has also been attempted by Matsumoto and Horide on ternary systems and below.…”

Section: Applicationmentioning

confidence: 99%

See 1 more Smart Citation

A Critical Review of Machine Learning of Energy Materials

Chen

Zuo

et al. 2020

Advanced Energy Materials

395

281

View full text Add to dashboard Cite

Machine learning (ML) is rapidly revolutionizing many fields and is starting to change landscapes for physics and chemistry. With its ability to solve complex tasks autonomously, ML is being exploited as a radically new way to help find material correlations, understand materials chemistry, and accelerate the discovery of materials. Here, an in‐depth review of the application of ML to energy materials, including rechargeable alkali‐ion batteries, photovoltaics, catalysts, thermoelectrics, piezoelectrics, and superconductors, is presented. A conceptual framework is first provided for ML in materials science, with a broad overview of different ML techniques as well as best practices. This is followed by a critical discussion of how ML is applied in energy materials. This review is concluded with the perspectives on major challenges and opportunities in this exciting field.

show abstract

Section: Applicationmentioning

confidence: 99%

“…projecting the periodic table to an image have been adopted by Zheng et al [348] to find novel X 2 YZ compounds. The prediction of T cr has also been attempted by Matsumoto and Horide [349] on ternary systems and below.…”

Section: Wwwadvancedsciencenewscommentioning

confidence: 99%

A Critical Review of Machine Learning of Energy Materials

Chen

Zuo

et al. 2020

Advanced Energy Materials

395

281

View full text Add to dashboard Cite

show abstract

“…This paradigm change is further promoted by projects like the materials genome initiative (Materials genome initiative) that aim to bridge the gap between experiment and theory and promote a more data-intensive and systematic research approach. A multitude of already successful machine learning applications in materials science can be found, e.g., the prediction of new stable materials, [27][28][29][30][31][32][33][34][35] the calculation of numerous material properties, [36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51] and the speeding up of firstprinciple calculations. 52 Machine learning algorithms have already revolutionized other fields, such as image recognition.…”

Section: Introductionmentioning

confidence: 99%

Recent advances and applications of machine learning in solid-state materials science

et al. 2019

View full text Add to dashboard Cite

One of the most exciting tools that have entered the material science toolbox in recent years is machine learning. This collection of statistical methods has already proved to be capable of considerably speeding up both fundamental and applied research. At present, we are witnessing an explosion of works that develop and apply machine learning to solid-state systems. We provide a comprehensive overview and analysis of the most recent research in this topic. As a starting point, we introduce machine learning principles, algorithms, descriptors, and databases in materials science. We continue with the description of different machine learning approaches for the discovery of stable materials and the prediction of their crystal structure. Then we discuss research in numerous quantitative structure-property relationships and various approaches for the replacement of first-principle methods by machine learning. We review how active learning and surrogate-based optimization can be applied to improve the rational design process and related examples of applications. Two major questions are always the interpretability of and the physical understanding gained from machine learning models. We consider therefore the different facets of interpretability and their importance in materials science. Finally, we propose solutions and future research paths for various challenges in computational materials science.

show abstract

“…Another instanceo ft he relevance of the structure as ap redictive tool is its recent use in the estimation of enthalpies of formation of several compounds using neuraln etworks and the order structure of the systems. [117] Likewise, machine learning methods can be used to estimate properties and there are already resultsw here learningi sd efined and developed based upon ordered hypergraphs. [118] The mathematicsofordered hypergraphs is rather new,i ts future developments mayb ring up more predictive tools for chemistry.W ee nvision this as af ruitful field of research with implications for the periodic systems.…”

Section: Bringing Back Predictionsmentioning

confidence: 99%

Challenges for the Periodic Systems of Elements: Chemical, Historical and Mathematical Perspectives

Restrepo

2019

Chemistry A European J

View full text Add to dashboard Cite

We celebrate 150 years of periodic systems that reached their maturity in the 1860s. They began as pedagogical efforts to project corpuseso fs ubstances on the similarity and order relationshipso ft he chemical elements. However,t hese elements are not the canned substances wrongly displayed in many periodic tables, but rather the abstract preserved entitiesi nc ompound transformations. We celebrate the systems, rather than their tables or ultimate table. The periodic law,w ea rgue, is not an all-encompassing achievement, as it does not apply to every property of all elements and compounds.P eriodic systemsh ave been generaliseda so rdered hypergraphs, which solves the longlasting question on the mathematical structureo ft he systems. In this essay,i ti ss hown that these hypergraphs may solve current issues such as order reversals in super-heavy elements and lacko fs ystem predictivep ower.W ed iscuss research in extending the limits of the systemsi nt he superheavy-atom region and draw attention to other limits:t he [a] Dr.

show abstract

Machine learning material properties from the periodic table using convolutional neural networks

Abstract: Convolutional neural networks directly learned chemical information from the periodic table to predict the enthalpy of formation and compound stability.

Cited by 90 publications

References 34 publications

A Critical Review of Machine Learning of Energy Materials

A Critical Review of Machine Learning of Energy Materials

Recent advances and applications of machine learning in solid-state materials science

Challenges for the Periodic Systems of Elements: Chemical, Historical and Mathematical Perspectives

Contact Info

Product

Resources

About