Compositional design of compounds with elements not in training data using supervised learning
)Graphical Abstract
Abstract
An issue of current interest in the use of machine learning models to predict compositions of materials is their reliability in predicting outcomes with elements not included in the training data. We show that the phase diagram of the ceramic (Ba1–x–yCaxSry)(Ti1–u–v–wZruSnvHfw)O3 can be accurately predicted if the feature values of unknown elements do not exceed the range of values for existing elements in the training data. In particular, we employ physical features as descriptors and compositions as weights to show that by excluding an element, such as Zr, Sn or Hf from the training set and treating it as an unknown element, the machine learning model accurately predicts the property only if the feature values of the unknown element does not exceed the range of values of existing elements in the training set. By adding a small amount of data for the unknown element restores the prediction accuracy. We demonstrate this for BaTiO3 ceramics doped with rare earth elements where the prediction accuracy is restored if the physical feature space is suitably enlarged with training data. The prediction error increases with the Euclidean distance of the testing sample relative to the nearest training sample in the physical feature space. Our work provides an effective strategy for extending machine learning models for material compositions beyond the scope of available data.
References
Gubernatis JE, Lookman T. Machine learning in materials design and discovery: examples from the present and suggestions for the future. Phys Rev Mater 2018;2:120301.
Tian Y, Yuan RH, Xue DZ, Zhou YM, Wang YF, Ding XD, et al. Determining multi-component phase diagrams with desired characteristics using active learning. Adv Sci 2021;8:2003165.
Li LL, Yang YD, Zhang DW, Ye ZG, Jesse S, Kalinin SV, et al. Machine learning–enabled identification of material phase transitions based on experimental data: exploring collective dynamics in ferroelectric relaxors. Sci Adv 2018;4:eaap8672.
Iwasaki Y, Kusne AG, Takeuchi I. Comparison of dissimilarity measures for cluster analysis of X-ray diffraction data from combinatorial libraries. npj Comput Mater 2017;3:4.
Pradhan DK, Kumari S, Strelcov E, Pradhan DK, Katiyar RS, Kalinin SV, et al. Reconstructing phase diagrams from local measurements via Gaussian processes: mapping the temperature-composition space to confidence. npj Comput Mater 2018;4:23.
Ghosh K, Stuke A, Todorović M, Jørgensen PB, Schmidt MN, Vehtari A, et al. Deep learning spectroscopy: neural networks for molecular excitation spectra. Adv Sci 2019;6:1801367.
Oviedo F, Ren ZK, Sun SJ, Settens C, Liu Z, Hartono NTP, et al. Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks. npj Comput Mater 2019;5:60.
Kou ZN, Hashemi A, Puska MJ, Krasheninnikov AV, Komsa H-P. Simulating Raman spectra by combining first-principles and empirical potential approaches with application to defective MoS2. npj Comput Mater 2020;6:59.
Ren CJ, Lu SH, Wu YL, Ouyang YX, Zhang YH, Li Q, et al. A universal descriptor for complicated interfacial effects on electrochemical reduction reactions. J Am Chem Soc 2022;144:12874–83.
He JJ, Li JJ, Liu CB, Wang CX, Zhang Y, Wen C, et al. Machine learning identified materials descriptors for ferroelectricity. Acta Mater 2021;209:116815.
Im J, Lee S, Ko T, Kim HW, Hyon Y, Chang H. Identifying Pb-free perovskites for solar cells by machine learning. npj Comput Mater 2019;5:1–8.
Kim C, Pilania G, Ramprasad R. From organized high-throughput data to phenomenological theory using machine learning: the example of dielectric breakdown. Chem Mater 2016;28:1304–11.
He JJ, Yu CY, Hou YX, Su XP, Li JJ, Liu CB, et al. Accelerated discovery of high-performance piezocatalyst in BaTiO3-based ceramics via machine learning. Nano Energy 2022;97:107218.
Gong J, Chu S, Mehta RK, Mcgaughey AJH. XGBoost model for electrocaloric temperature change prediction in ceramics. npj Comput Mater 2022;8:140.
Stupp D, Sharon E, Bloch I, Zitnik M, Zuk O, Tabach Y. Co-evolution based machine-learning for predicting functional interactions between human genes. Nat Commun 2021;12:6454.
Kunkel C, Margraf JT, Chen K, Oberhofer H, Reuter K. Active discovery of organic semiconductors. Nat Commun 2021;12:2422.
Jia HH, Ju DY, Cao JT. Machine learning based optimization method for vacuum carburizing process and its application. J Mater Inf 2023;3:9.
Ma BW, Yu FY, Zhou P, Wu X, Zhao CL, Lin C, et al. Machine learning accelerated discovery of high transmittance in (K0.5Na0.5)NbO3-based ceramics. J Mater Inf 2023;3:13.
Gao TC, Gao JB, Zhang JL, Song B, Zhang LJ. Development of an accurate “composition-process-properties” dataset for SLMed Al-Si-(Mg) alloys and its application in alloy design. J Mater Inf 2023;3:6.
Liu Y, Yang ZW, Zou XX, Ma SC, Liu DH, Avdeev M, et al. Data quantity governance for machine learning in materials science. Natl Sci Rev 2023;10:nwad125.
Liu Y, Wang SY, Yang ZW, Avdeev M, Shi SQ. Auto-MatRegressor: liberating machine learning alchemists. Sci Bull 2023;68:1259–70.
Wen C, Zhang Y, Wang CX, Xue DZ, Bai Y, Antonov S, et al. Machine learning assisted design of high entropy alloys with desired property. Acta Mater 2019;170:109–17.
Yuan RH, Xue DQ, Xue DZ, Li JS, Ding XD, Sun J, et al. Knowledge-based descriptor for the compositional dependence of the phase transition in BaTiO3-based ferroelectrics. ACS Appl Mater Interfaces 2020;12:44970–80.
Wang CX, Wang XX, Zhang TY, Qian P, Lookman T, Su YJ. A descriptor for the design of 2D MXene hydrogen evolution reaction electrocatalysts. J Mater Chem A 2022;10:18195–205.
Bartel CJ, Sutton C, Goldsmith BR, Ouyang R, Musgrave CB, Ghiringhelli LM, et al. New tolerance factor to predict the stability of perovskite oxides and halides. Sci Adv 2019;5:eaav0693.
Weng BC, Song ZL, Zhu RL, Yan QY, Sun QD, Grice CG, et al. Simple descriptor derived from symbolic regression accelerating the discovery of new perovskite catalysts. Nat Commun 2020;11:3513.
Yuan RH, Liu Z, Balachandran PV, Xue DQ, Zhou YM, Ding XD, et al. Accelerated discovery of large electrostrains in BaTiO3-based piezoelectrics using active learning. Adv Mater 2018;30:1702884.
Xue DZ, Balachandran PV, Hogden J, Theiler J, Xue DQ, Lookman T. Accelerated search for materials with targeted properties by adaptive design. Nat Commun 2016;7:11241.
Liu P, Huang HY, Antonov S, Wen C, Xue DZ, Chen HW, et al. Machine learning assisted design of γ′-strengthened Co-base superalloys with multi-performance optimization. npj Comput Mater 2020;6:62.
Rickman JM, Chan HM, Harmer MP, Smeltzer JA, Marvel CJ, Roy A, et al. Materials informatics for the screening of multi-principal elements and high-entropy alloys. Nat Commun 2019;10:2618.
Balachandran PV, Kowalski B, Sehirlioglu A, Lookman T. Experimental search for high-temperature ferroelectric perovskites guided by two-step machine learning. Nat Commun 2018;9:1668.
Li ZZ, Xu QC, Sun QD, Hou ZF, Yin W-J. Thermodynamic stability landscape of halide double perovskites via high-throughput computing and machine learning. Adv Funct Mater 2019;29:1807280.
He JJ, Su XP, Wang CX, Li JJ, Hou YX, Li ZH, et al. Machine learning assisted predictions of multi-component phase diagrams and fine boundary information. Acta Mater 2022;240:118341.
Zhao CL, Wang H, Xiong J, Wu JG. Composition-driven phase boundary and electrical properties in (Ba0.94Ca0.06)(Ti1–xMx)O3 (M = Sn, Hf, Zr) lead-free ceramics. Dalton Trans 2016;45:6466–80.
Zhao CL, Feng YM, Wu HP, Wu JG. Phase boundary design and high piezoelectric activity in (1−x)(Ba0.93Ca0.07)TiO3-xBa(Sn1–yHfy)O3 lead-free ceramics. J Alloys Compd 2016;666:372–9.
Liu WF, Ren XB. Large piezoelectric effect in Pb-free ceramics. Phys Rev Lett 2009;103:257602.
Bao HX, Zhou C, Xue DZ, Gao JH, Ren XB. A modified lead-free piezoelectric BZT–xBCT system with higher TC. J Phys D Appl Phys 2010;43:465401.
Yao YG, Zhou C, Lv DC, Wang D, Wu HJ, Yang YD, et al. Large piezoelectricity and dielectric permittivity in BaTiO3-xBaSnO3 system: the role of phase coexisting. EPL Europhys Lett 2012;98:27008.
Han F, Bai Y, Qiao LJ, Guo D. A systematic modification of the large electrocaloric effect within a broad temperature range in rare-earth doped BaTiO3 ceramics. J Mater Chem C 2016;4:1842–9.
Wu Y-J, Fang L, Xu YB. Predicting interfacial thermal resistance by machine learning. npj Comput Mater 2019;5:56.
京公网安备11010802044758号