Abstract: Materials are the foundation of the national economy, the discovery of new materials gives impetus to the development of modern science and technological innovation. The traditional “trial and error” experimental methods are no longer applicable to the research of modern materials owing to the disadvantages of high cost, long period and great contingency. In recent years, with the rapid development of artificial intelligence and data-driven approach, as a main branch and an important tool of them, machine learning is receiving increasing attention and showing tremendous potential. The integration of machine learning into material science research can greatly improve the precision and efficiency, and provide new opportunities for the research and development of new materials and the study of the basic theory. Machine learning technology combines knowledge of computer science, probability theory, statistics, database theory and engineering. It shows a faster computing speed and good generalization ability, and can effectively deal with some systems and problems difficult to tackle by traditional experiments and numerical simulation. In the past decade, the applications of machine learning in material science research have shown explosive growth,especially in the synthesis and design of new materials, the property prediction, characterization of the microstructure, and the improvement of material calculation and simulation methods. Machine learning will be indispensable in the development of material science and engineering in the future. At present, how to obtain a large number of experimental data and build effective data set is still a hot spot and difficulty in the application of machine learning in the field of material science. This paper outlines the basic principles, workflows and common algorithms of machine learning, briefly describes the research focus and application progress of machine learning technology in the field of materials science, and analyzes the existing problems of machine learning in mate-rials science research. Meanwhile, some hot spots of the material field in the future are pointed out.
米晓希, 汤爱涛, 朱雨晨, 康靓, 潘复生. 机器学习技术在材料科学领域中的应用进展[J]. 材料导报, 2021, 35(15): 15115-15124.
MI Xiaoxi, TANG Aitao, ZHU Yuchen, KANG Jing, PAN Fusheng. Research Progress of Machine Learning in Material Science. Materials Reports, 2021, 35(15): 15115-15124.
1 Jordan M I, Mitchell T M. Science,2015,349,255. 2 Butler K T, Davies D W, Cartwright H, et al. Nature,2018,559,547. 3 Liu H, Xu C, Liang J. Physics of Life Reviews,2017,21,233. 4 Cambria E, White B. Computational Intelligence Magazine IEEE,2014,9(2),48. 5 Shin H C, Roth H R, Gao M, et al. IEEE Transactions on Medical Imaging,2016,35(5),1285. 6 Criminisi A. Medical Image Analysis,2016,33,91. 7 Yang K K, Wu Z, Arnold F H. Nature Methods,2019,16(8),687. 8 Sean E, Ana C, Kimberley M,et al. Nature Materials,2019,18(5),435. 9 Cully A, Clune J, Tarapore D, et al. Nature,2014,521(7553),503. 10 Schmidt J, Marques M, Botti S, et al. npj Computational Materials,2019,5,83. 11 Feng N, Wang H J, Li M. Information Sciences,2014,256,57. 12 Ramprasad R, Batra R, Pilania G, et al. npj Computational Materials,2017,3(1),54. 13 Jain A, Hautier G, Ong S P, et al. Journal of Materials Research,2016,31(8),977. 14 Yu H K. International Statistical Review,2001,69(3),385. 15 Shakhnarovich G, Darrell T, Indyk P. Nearest-neighbor methods in lear-ning and vision: theory and practice, MIT Press, Boston, 2005. 16 Burges C J. A tutorial on support vector machines for pattern recognition, Kluwer Academic Publishers, Dordrecht,1998. 17 Safavian S R, Landgrebe D. IEEE Transactions on Systems Man & Cybernetics,2002,21(3),660. 18 Zhu J F. Design and application of greenhouse environmental regulation rules based on decision tree. Master’s Thesis, Zhejiang University, China,2017(in Chinese). 朱剑峰.基于决策树的温室环境调控规则设计及其应用研究.硕士学位论文,浙江大学,2017. 19 Dempster A P, Laird N M, Rubin D B. Journal of the Royal Statistical Societ,Series B: Methodological,1977,39(1),1. 20 Goh G B, Hodas N O, Vishnu A. Journal of Computational Chemistry,2017,38(16),1291. 21 Lecun Y, Bengio Y, Hinton G. Nature,2015,521(7553),436. 22 Wu W, Sun Q. Scientia sinica Physica, Mechanica & Astronomica,2018,48(10),58(in Chinese). 吴炜,孙强.中国科学:物理学力学天文学,2018,48(10),58. 23 Liu Y, Zhao T, Ju W, et al. Journal of Materiomics,2017,3(3),159. 24 Lin Q Y. Journal of Chongqing University of Technology (Natural Science),2019,33(10),121(in Chinese). 林倩瑜.重庆理工大学学报(自然科学),2019,33(10),121. 25 Aritra C, Elizabeth K, Bülent Y, et al. Computational Materials Science,2016,123,176. 26 Corey E J, Wipke W T. Science,1969,166(3902),178. 27 Zhang H, Li N, Li K, et al. Acta Crystallographica,2010,63(6),812. 28 Li C, Lu X, Ding W, et al. Acta Crystallographica,2010,64(6),702. 29 Xu Q, Li Z, Liu M, et al. Journal of Physical Chemistry Letters,2018,9,6948. 30 Sun Z T, Li Z Z, Cheng G J, et al. Chinese Science Bulletin,2019(32),3270(in Chinese). 孙中体,李珍珠,程观剑,等.科学通报,2019(32),3270. 31 Raccuglia P, Elbert K C, Adler P D, et al. Nature,2016,533,73. 32 Kumar J N, Li Q, Tang K Y, et al. npj Computational Materials,2019,5,73. 33 Pillong M, Marx C, Piechon P, et al. CrystEngComm,2017,19,3711. 34 Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative Adversarial Networks,2014,3,2672. 35 Attarian Shandiz M, Gauvin R. Computational Materials Science,2016,117,270. 36 Fujimura K, Seko A, Koyama Y, et al. Advanced Energy Materials,2013,3(8),980. 37 Jie J S, Hu Z X,Qian G Y, et al. Science Bulletin,2019,64(9),612. 38 Gu G X, Chen C T, Buehler M J. Extreme Mechanics Letters,2017,18,19. 39 Zeng S, Zhao Y, Li G, et al. npj Computational Materials,2019,5,84. 40 Zhou Z, Zhou Y, He Q, et al. npj Computational Materials,2019,5,128. 41 Liu B. Development and research of a magnesium alloy expert system. Ph. D. Thesis, Chongqing University, China,2011(in Chinese). 刘彬.镁合金专家系统的开发研究.博士学位论文,重庆大学,2011. 42 Wang C, Fu H, Jiang L, et al. npj Computational Materials,2019,5,87. 43 Tagade P M, Adiga S P, Pandian S, et al. npj Computational Materials,2019,5,127. 44 Liu J, Mohan A, Kalia R K, et al. Computational Materials Science,2020,173,15. 45 Pokuri B S, Ghosal S, Kokate A, et al. npj Computational Materials,2019,5,95. 46 Ament S E, Stein H S, Guevarra D, et al. npj Computational Materials,2019,5,77. 47 Saito Y, Shin K, Terayama K, et al. npj Computational Materials,2019,5,124. 48 Chan H, Cherukara M, Loeffler T D. npj Computational Materials,2020,6,1. 49 Orme A D, Chelladurai I, Rampton T M, et al. Computational Materials Science,2016,124,353. 50 Tong Z, Wang L, Zhu G, et al. Metallurgical and Materials Transactions A,2019,35(9),2719. 51 Gao Q. Study on defect detection of rolled sheet for magnesium alloy based on deep learning. Master’s Thesis,University of Science and Technology Liaoning, China,2018(in Chinese). 高青.基于深度学习的镁合金轧制薄板缺陷检测问题的研究.硕士学位论文,辽宁科技大学,2018. 52 Attarian Shandiz M, Gauvin R. Computational Materials Science,2016,117,270. 53 Terentyev D, Bonny G, Castin N, et al. Journal of Nuclear Materials,2011,409(2),167. 54 Shuaihua L, Qiong H. Nature Communications,2018,9,1. 55 Bernstein N, Csanyi G, Deringer V L, et al. npj Computational Mate-rials,2019,5(1),99. 56 Hongxiang Z, Ghanshyam P, Xiangdong D, et al. npj Computational Materials,2018,4(1),48. 57 Cooper A, Kstner J, Urban A, et al. npj Computational Materials,2020,6,54. 58 Tian J, Zhu D J, Yang W H,et al. Computer Science,2018,45(11A),58(in Chinese). 田娟,朱定局,杨文翰,等.计算机科学,2018,45(11A),58. 59 Liu R, Zhang N. Computer Science,2021,48(7),137(in Chinese). 刘荣,张宁.计算机科学,2021,48(7),137. 60 Belsky A, Hellenbrandt M, Karen V L, et al. Acta Crystallographica,2010,58(3),364. 61 Allen F H. Acta Crystallographica Section B Structural Science,2002,58,380. 62 Kirklin S, Saal J E, Meredig B, et al. npj Computational Materials,2015,1,15010. 63 Curtarolo S, Setyawan W, Hart G L W, et al. Computational Materials Science,2012,58,218. 64 Shaikhina T, Khovanova N A. Artificial Intelligence in Medicine,2017,75,51.