Multi-fidelity learning in materials informatics: methodologies, applications, and outlook

Abstract

Data-driven methods are transforming materials design by accelerating the discovery of new compounds and the optimization of existing systems. However, the progress of such approaches is often constrained by the scarcity of high-fidelity data from experiments and advanced simulations. Multi-fidelity (MF) learning has emerged as a powerful strategy to address this challenge by integrating information from diverse data sources that vary in accuracy and cost. In this review, we provide a systematic overview of the major methodologies for MF learning, including statistical and parametric models, machine learning models with fidelity features, correction-based models such as co-kriging, deep learning frameworks, and active learning frameworks. We discuss the strengths, limitations, and typical applications of each method in materials science, with illustrative examples spanning electronic structure modeling, alloy design, and interatomic potential development. Cross-cutting issues are also examined, including the bias-variance trade-off, data requirements for nested vs. non-nested designs, and computational scalability. Finally, we highlight outstanding challenges and outline emerging opportunities, such as physics-informed and generative MF models, standardized datasets, and integration with autonomous laboratories. Together, these perspectives define a roadmap for advancing MF learning as a core enabler of next-generation materials discovery.

Keywords

Multi-fidelity learning / materials informatics / surrogate modeling / active learning / data-driven materials design

References

[1] Batra R,Ramprasad R.Emerging materials intelligence ecosystems propelled by machine learning.Nat Rev Mater2020;6:655-78

[2] Rao Z,Xie R.Machine learning-enabled high-entropy alloy discovery.Science2022;378:78-85

[3] Sohail Y,Xue D.Machine-learning design of ductile FeNiCoAlTa alloys with high strength.Nature2025;643:119-24 PMCID:PMC12221976

[4] Wang T,Martins ML.Machine-learning-assisted material discovery of oxygen-rich highly porous carbon active materials for aqueous supercapacitors.Nat Commun.2023;14:4607 PMCID:PMC10393944

[5] Wu J,Hu M.Inverse design workflow discovers hole-transport materials tailored for perovskite solar cells.Science.2024;386:1256-64

[6] Tian Y,Dang P,Zhou Y.Noise-aware active learning to develop high-temperature shape memory alloys with large latent heat.Adv Sci2024;11:e2406216 PMCID:11600200

[7] Xian Y,Tian Y.Compositional design of multicomponent alloys using reinforcement learning.Acta Mater2024;274:120017

[8] Wei Q,Yuan H.Divide and conquer: machine learning accelerated design of lead-free solder alloys with high strength and high ductility.npj Comput Mater2023;9:201

[9] Gong X,Duan W.Generalizing deep learning electronic structure calculation to the plane-wave basis.Nat Comput Sci2024;4:752-60 PMCID:PMC11499277

[10] Dang L,Tang D,Wu B.A fatigue life prediction framework of laser-directed energy deposition Ti-6Al-4V based on physics-informed neural network.IJSI2025;16:327-54

[11] Rasul A,Perumal V,Iqbal M.An artificial neural network model for determining stress concentration factors for fatigue design of tubular T-joint under compressive loads.IJSI2024;15:633-52

[12] Yuan R,Li J.Machine learning-enabled design of ferroelectrics with multiple properties via a Landau model.Acta Mater2025;286:120760

[13] Wang W,Tian S.Automated pipeline for superalloy data by text mining.npj Comput Mater2022;8:9

[14] Butler KT,Cartwright H,Walsh A.Machine learning for molecular and materials science.Nature2018;559:547-55

[15] Raccuglia P,Adler PD.Machine-learning-assisted materials discovery using failed experiments.Nature2016;533:73-6

[16] Dinic F,Dong T.Applied machine learning for developing next‐generation functional materials.Adv Funct Mater2021;31:2104195

[17] Rickman J,Kalinin S.Materials informatics: from the atomic-level to the continuum.Acta Mater2019;168:473-510

[18] Hong T,Jin D.Discovery of new topological insulators and semimetals using deep generative models.npj Quantum Mater2025;10:12

[19] Xie J.Prospects of materials genome engineering frontiers.Mater Genome Eng Adv2023;1:e17

[20] Jiang X,Bai Y.Interpretable machine learning applications: a promising prospect of AI for materials.Adv Funct Mater2025;35:2507734

[21] Hart GLW,Toher C.Machine learning for alloys.Nat Rev Mater2021;6:730-55

[22] Hippalgaonkar K,Wang X,Kirkpatrick J.Knowledge-integrated machine learning for materials: lessons from gameplaying and robotics.Nat Rev Mater2023;8:241-60

[23] Tabor DP,Saikin SK.Accelerating the discovery of materials for clean energy in the era of smart automation.Nat Rev Mater2018;3:5-20

[24] Gong X,Zou N,Duan W.General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian.Nat Commun2023;14:2848 PMCID:PMC10199065

[25] Li H,Zou N.Deep-learning density functional theory Hamiltonian for efficient ab initio electronic-structure calculation.Nat Comput Sci2022;2:367-77 PMCID:PMC11499279

[26] Cohen AJ,Yang W.Challenges for density functional theory.Chem Rev2012;112:289-320

[27] Janesko BG.Replacing hybrid density functional theory: motivation and recent advances.Chem Soc Rev2021;50:8470-95

[28] Takeno S,Fukuoka H,Shiga M.Cost-effective search for lower-error region in material parameter space using multifidelity Gaussian process modeling.Phys Rev Materials2020;4:083802

[29] Chen C,Ye W,Ong SP.Learning properties of ordered and disordered materials from multi-fidelity data.Nat Comput Sci2021;1:46-53

[30] Lu L,Kumar P,Karniadakis GE.Extraction of mechanical properties of materials through deep learning from instrumented indentation.Proc Natl Acad Sci U S A2020;117:7052-62 PMCID:PMC7132259

[31] Jain A,Moore CJ.A high-throughput infrastructure for density functional theory calculations.Comput Mater Sci2011;50:2295-310

[32] Pilania G,Lookman T.Multi-fidelity machine learning models for accurate bandgap predictions of solids.Comput Mater Sci2017;129:156-63

[33] Egorova O,Woods DC.Multifidelity statistical machine learning for molecular crystal structure prediction.J Phys Chem A2020;124:8065-78

[34] Batra R,Uberuaga BP.Multifidelity information fusion with machine learning: a case study of dopant formation energies in hafnia.ACS Appl Mater Interfaces2019;11:24906-18

[35] Jha D,Tavazza F.Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning.Nat Commun2019;10:5316 PMCID:PMC6874674

[36] Peng B,Ye D.DIC/DSI based studies on the local mechanical behaviors of HR3C/T92 dissimilar welded joint during plastic deformation.Mater Sci Eng A2022;857:144073

[37] Valladares H,Zhu L.Gaussian process-based prognostics of lithium-ion batteries and design optimization of cathode active materials.J Power Sources2022;528:231026

[38] Lindsey RK,Hamel S,Goldman N.ChIMES carbon 2.0: a transferable machine-learned interatomic model harnessing multifidelity training data.npj Comput Mater2025;11:26

[39] Ko TW.Data-efficient construction of high-fidelity graph deep learning interatomic potentials.npj Comput Mater2025;11:65

[40] Kim J,Kim J.Data-efficient multifidelity training for high-fidelity machine learning interatomic potentials.J Am Chem Soc2025;147:1042-54

[41] Perdikaris P,Damianou A,Karniadakis GE.Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling.Proc Math Phys Eng Sci2017;473:20160751 PMCID:PMC5332612

[42] Liu M,Hegde VI,Wolverton C.High-throughput hybrid-functional DFT calculations of bandgaps and formation energies and multifidelity learning with uncertainty quantification.Phys Rev Materials2024;8:043803

[43] Yang Z,Huang L.Machine learning-based extraction of mechanical properties from multi-fidelity small punch test data.Adv Manuf2025;13:511-24

[44] Yi J,Bessa MA.Single-to-multi-fidelity history-dependent learning with uncertainty quantification and disentanglement: application to data-driven constitutive modeling.Comput Methods Appl Mech Eng2026;448:118479

[45] Yi, J.; Cheng, J.; Bessa, M. A. Practical multi-fidelity machine learning: fusion of deterministic and Bayesian models. arXiv 2025, arXiv:2407.15110. Available online: https://arxiv.org/abs/2407.15110 (accessed 26 March 2026).

[46] Cheng M,Hu J,Zhou Q.A multi-fidelity surrogate modeling method based on variance-weighted sum for the fusion of multiple non-hierarchical low-fidelity data.Struct Multidisc Optim2021;64:3797-818

[47] Palizhati A,Aykol M,Hummelshøj JS.Agents for sequential learning using multiple-fidelity data.Sci Rep2022;12:4694 PMCID:PMC8933401

[48] Le Gratiet, L. Multi-fidelity Gaussian process regression for computer experiments. Ph.D. Thesis, Université Paris-Diderot - Paris VII, Paris, France, 2013. https://theses.hal.science/tel-00866770/file/manuscrit.pdf (accessed 2026-3-26).

[49] Meng X.A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems.J Comput Phys2020;401:109020

[50] Moss, H. B.; Leslie, D. S.; Rayson, P. MUMBO: MUlti-task max-value Bayesian optimization. In Machine Learning and Knowledge Discovery in Databases; Hutter, F., Kersting, K., Lijffijt, J., Valera, I., Eds.; Springer International Publishing: Cham, 2021; pp 447-62.

[51] Xu X,Hu Y.Discovery of thermosetting polymers with low hygroscopicity, low thermal expansivity, and high modulus by machine learning.J Mater Chem A2023;11:12918-27

[52] Patra A,Chandrasekaran A,Huan TD.A multi-fidelity information-fusion approach to machine learn and predict polymer bandgap.Comput Mater Sci2020;172:109286

[53] Molkeri A,Couperthwaite R.On the importance of microstructure information in materials design: PSP vs PP.Acta Mater2022;223:117471

[54] Fare C,Benatan M,Pyzer-knapp EO.A multi-fidelity machine learning approach to high throughput materials screening.npj Comput Mater2022;8:257

[55] He Y,Allendorf MD.Metallic metal-organic frameworks predicted by the combination of machine learning methods and Ab initio calculations.J Phys Chem Lett2018;9:4562-9

[56] Yang J,Mannodi-Kanakkithodi A.Discovering novel halide perovskite alloys using multi-fidelity machine learning and genetic algorithm.J Chem Phys2024;160:064114

[57] Xu B,Lan H.Phthalonitrile melting point prediction enabled by multi-fidelity learning.J Mater Inf2024;4:21

[58] He G,Yin Z.Active learning inspired multi-fidelity probabilistic modelling of geomaterial property.Comput Methods Appl Mech Eng2024;432:117373

[59] Wang L,Wu B,Luo C.Multi-fidelity physics-informed machine learning framework for fatigue life prediction of additive manufactured materials.Comput Methods Appl Mech Eng2025;439:117924

[60] Sabanza-Gil V,Pacheco Gutiérrez D.Best practices for multi-fidelity Bayesian optimization in materials and molecular research.Nat Comput Sci2025;5:572-81

[61] Weisberg, S. Nonlinear regression. In Wiley Series in Probability and Statistics; Wiley, 2005.

[62] Bishop CM. Pattern recognition and machine learning. Information Science and Statistics. Springer; 2006. https://books.google.com/books/about/Pattern_Recognition_and_Machine_Learning.html?id=kTNoQgAACAAJ (accessed 2026-03-26).

[63] Yang B,Liu Y.Gaussian process fusion method for multi-fidelity data with heterogeneity distribution in aerospace vehicle flight dynamics.Eng Appl Artif Intell2024;138:109228

[64] Sella, V.; Pham, J.; Chaudhuri, A.; Willcox, K. E. Projection-based multifidelity linear regression for data-poor applications. In AIAA SCITECH 2023 Forum; National Harbor, MD & Online; Reston, VA: American Institute of Aeronautics and Astronautics, 2023.

[65] Ozbayram O,Graham-brady L.Heteroscedastic Gaussian process regression for material structure-property relationship modeling.Comput Methods Appl Mech Eng2024;431:117326

[66] Ibrahim A.Prediction of frequency-dependent optical spectrum for solid materials: a multioutput and multifidelity machine learning approach.ACS Appl Mater Interfaces2024;16:41145-56

[67] Mora C,Johnson T,Bostanabad R.Probabilistic neural data fusion for learning from an arbitrary number of multi-fidelity data sets.Comput Methods Appl Mech Eng2023;415:116207

[68] Zanjani Foumani Z, Shishehbor M, Yousefpour A, Bostanabad R. Multi-fidelity cost-aware Bayesian optimization.Comput Methods Appl Mech Eng2023;407:115937

[69] Toal DJJ.Some considerations regarding the use of multi-fidelity Kriging in the construction of surrogate models.Struct Multidisc Optim2014;51:1223-45

[70] Gratiet LL.Recursive Co-Kriging model for design of computer experiments with multiple levels of fidelity.Int J Uncertain Quant2014;4:365-86

[71] Forrester AI,Keane AJ.Multi-fidelity optimization via surrogate modelling.Proc R Soc A2007;463:3251-69

[72] Xiao M,Breitkopf P,Zhang W.Extended Co-Kriging interpolation method based on multi-fidelity data.Appl Math Comput2018;323:120-31

[73] Lu CK.Conditional deep gaussian processes: multi-fidelity kernel learning.Entropy2021;23:1545 PMCID:PMC8625033

[74] Cutajar, K.; Pullin, M.; Damianou, A.; Lawrence, N.; González, J. Deep Gaussian processes for multi-fidelity modeling. arXiv 2019, arXiv:1903.07320. Available online: https://arxiv.org/abs/1903.07320 (accessed 26 March 2026).

[75] Raissi, M.; Karniadakis, G. Deep multi-fidelity Gaussian processes. arXiv 2016, arXiv:1604.07484. Available online: https://arxiv.org/abs/1604.07484 (accessed 26 March 2026).

[76] Song X,Sun W.A radial basis function-based multi-fidelity surrogate model: exploring correlation between high-fidelity and low-fidelity models.Struct Multidisc Optim2019;60:965-81

[77] Acar P.Crystal plasticity model calibration for Ti-7Al alloy with a multi-fidelity computational scheme.Integr Mater Manuf Innov2018;7:186-94

[78] Shah AA,Xing WW.Rapid high-fidelity quantum simulations using multi-step nonlinear autoregression and graph embeddings.npj Comput Mater2025;11:57

[79] Pan W,Guo C.An efficient/accurate multi-scale fatigue prediction method for Metal-Polymer hybrid (MPH) interface.Int J Fatigue2024;184:108304

[80] Saunders RN,Elwany A,Lagoudas D.Metal AM process-structure-property relational linkages using Gaussian process surrogates.Additive Manuf2023;62:103398

[81] Morse L,Aliabadi M.A multi-fidelity modelling approach to the statistical inference of the equivalent initial flaw size distribution for multiple-site damage.Int J Fatigue2019;120:329-41

[82] Tran A,Wildey T.Multi-fidelity machine-learning with uncertainty quantification and Bayesian optimization for materials design: application to ternary random alloys.J Chem Phys2020;153:074705

[83] Menon N,Basak A.Multi-fidelity surrogate-based process mapping with uncertainty quantification in laser directed energy deposition.Materials2022;15:2902 PMCID:PMC9025529

[84] Wiangkham A,Kasemsri R,Tantrairatn S.Improvement of mixed-mode I/II fracture toughness modeling prediction performance by using a multi-fidelity surrogate model based on fracture criteria.Materials2022;15:8580 PMCID:PMC9740513

[85] Howard AA,Karniadakis GE.Multifidelity deep operator networks for data-driven and physics-informed problems.J Comput Phys2023;493:112462

[86] Motamed, M. A multi-fidelity neural network surrogate sampling method for uncertainty quantification. arXiv 2020, arXiv:1909.01859. Available online: https://arxiv.org/abs/1909.01859 (accessed 26 March 2026).

[87] Song, D. H.; Tartakovsky, D. M. Transfer Learning on Multi-Fidelity Data. arXiv 2021, arXiv:2105.00856. Available online: https://arxiv.org/abs/2105.00856 (accessed 26 March 2026).

[88] Chakraborty S.Transfer learning based multi-fidelity physics informed deep neural network.J Comput Phys2021;426:109942

[89] Snapp KL,Pang R.A physics-informed impact model refined by multi-fidelity transfer learning.Extreme Mechanics Letters2024;72:102223

[90] Chen D,Liu K.A physics-informed neural network approach to fatigue life prediction using small quantity of samples.Int J Fatigue2023;166:107270

[91] Islam M,Mojumder S.Extraction of material properties through multi-fidelity deep learning from molecular dynamics simulation.Comput Mater Sci2021;188:110187

[92] Liu B,Wang G.A method for calculating residual strength of crack arrest hole on tungsten-copper functionally graded materials by phase-field gradient element combined with multi-fidelity neural network.Materials2025;18:1973 PMCID:PMC12072201

[93] De Breuck P, Heymans G, Rignanese G. Accurate experimental band gap predictions with multifidelity correction learning.J Mater Inf2022;2:10

[94] Cleeman J,Nastarowicz E.Partial-physics-informed multi-fidelity modeling of manufacturing processes.J Mater Process Technol2023;320:118125

[95] Khatamsaz D,Couperthwaite R.Adaptive active subspace-based efficient multifidelity materials design.Mater Des2021;209:110001

[96] Hernández-García, A.; Saxena, N.; Jain, M.; Liu, C. H.; Bengio, Y. Multi-Fidelity Active Learning with GFlowNets. 2024. https://openreview.net/forum?id=dLaazW9zuF (accessed 2026-03-26).

[97] Koutsourelakis P.Accurate uncertainty quantification using inaccurate computational models.SIAM J Sci Comput2009;31:3274-300

[98] Wang Z,Chen H,He Y.Exploring multi-fidelity data in materials science: challenges, applications, and optimized learning strategies.Applied Sciences2023;13:13176

[99] Greenman KP,Gómez-Bombarelli R.Multi-fidelity prediction of molecular optical peaks with deep learning.Chem Sci2022;13:1152-62 PMCID:PMC8790778

[100] Burdick RK,Montgomery DC.A review of methods for measurement systems capability analysis.J Qual Technol2018;35:342-54

[101] Giannoukou, K.; Marelli, S.; Sudret, B. Uncertainty-aware multi-fidelity surrogate modeling with noisy data. arXiv 2024, arXiv:2401.06447. Available online: https://arxiv.org/abs/2401.06447 (accessed 26 March 2026).

[102] Konomi, B. A.; Karagiannis, G. Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: application to the WRF model. arXiv 2019, arXiv:1910.08063. Available online: https://arxiv.org/abs/1910.08063 (accessed 26 March 2026).

[103] Zhou Q,Guo Z,Jin P.A generalized hierarchical co-Kriging model for multi-fidelity data fusion.Struct Multidisc Optim2020;62:1885-904

[104] Rasmussen, CE.; Williams, CKI. Gaussian processes for machine learning; The MIT Press, 2005.

[105] Maldonado S.Quasi-two-layer morphodynamic model for bedload-dominated problems: bed slope-induced morphological diffusion.R Soc Open Sci2018;5:172018 PMCID:PMC5830787