Explainable machine learning in materials science
Source abstract
Abstract
Abstract Machine learning models are increasingly used in materials studies because of their exceptional accuracy. However, the most accurate machine learning models are usually difficult to explain. Remedies to this problem lie in explainable artificial intelligence (XAI), an emerging research field that addresses the explainability of complicated machine learning models like deep neural networks (DNNs). This article attempts to provide an entry point to XAI for materials scientists. Concepts are defined to clarify what explain means in the context of materials science. Example works are reviewed to show how XAI helps materials science research. Challenges and opportunities are also discussed.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · Explainable Artificial Intelligence (XAI) · Computational Drug Discovery Methods
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (100+)
Show source-provided references
1. Xu, Y. et al. High-throughput calculations of magnetic topological materials. Nature 586, 702–707 (2020).10.1038/s41586-020-2837-0 ↗
2. Ophus, C. Four-dimensional scanning transmission electron microscopy (4d-stem): From scanning nanodiffraction to ptychography and beyond. Microsc. Microanal. 25, 563–582 (2019).10.1017/S1431927619000497 ↗
3. Ludwig, A. Discovery of new materials using combinatorial synthesis and high-throughput characterization of thin-film materials libraries combined with computational methods. npj Comput. Mater. 5, 70 (2019).10.1038/s41524-019-0205-0 ↗
4. Behler, J. Perspective: machine learning potentials for atomistic simulations. J. Chem. Phys. 145, 170901 (2016).10.1063/1.4966192 ↗
5. Li, Z., Kermode, J. R. & De Vita, A. Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces. Phys. Rev. Lett. 114, 096405 (2015).10.1103/PhysRevLett.114.096405 ↗
6. Schmidt, J., Marques, M. R., Botti, S. & Marques, M. A. Recent advances and applications of machine learning in solid-state materials science. npj Comput. Mater. 5, 83 (2019).10.1038/s41524-019-0221-0 ↗
7. Schütt, K. T. et al. How to represent crystal structures for machine learning: Towards fast prediction of electronic properties. Phys. Rev. B 89, 205118 (2014).10.1103/PhysRevB.89.205118 ↗
8. Yang, Z. et al. Establishing structure-property localization linkages for elastic deformation of three-dimensional high contrast composites using deep learning approaches. Acta Mater. 166, 335–345 (2019).10.1016/j.actamat.2018.12.045 ↗
9. Gallagher, B. et al. Predicting compressive strength of consolidated molecular solids using computer vision and deep learning. Mater. Des. 190, 108541 (2020).10.1016/j.matdes.2020.108541 ↗
10. Zhang, H. et al. Dramatically enhanced combination of ultimate tensile strength and electric conductivity of alloys via machine learning screening. Acta Mater. 200, 803–810 (2020).10.1016/j.actamat.2020.09.068 ↗
11. Lu, Z. et al. Interpretable machine-learning strategy for soft-magnetic property and thermal stability in fe-based metallic glasses. npj Comput. Mater. 6, 187 (2020).10.1038/s41524-020-00460-x ↗
12. Adadi, A. & Berrada, M. Peeking inside the black-box: a survey on explainable artificial intelligence (xai). IEEE Access 6, 52138–52160 (2018).10.1109/ACCESS.2018.2870052 ↗
13. Montavon, G., Samek, W. & Müller, K.-R. Methods for interpreting and understanding deep neural networks. Digit. Signal Process. 73, 1–15 (2018).10.1016/j.dsp.2017.10.011 ↗
14. Nguyen, A., Yosinski, J. & Clune, J. Deep neural networks are easily fooled: High confidence predictions for unrecognizable images. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 427–436 (2015).10.1109/CVPR.2015.7298640 ↗
15. Ilyas, A. et al. Adversarial examples are not bugs, they are features. In Advances in Neural Information Processing Systems (NIPS), vol. 32 (2019).
16. Liu, Y., Wu, J.-M., Avdeev, M. & Shi, S.-Q. Multi-layer feature selection incorporating weighted score-based expert knowledge toward modeling materials with targeted properties. Adv. Theory Simul. 3, 1900215 (2020).10.1002/adts.201900215 ↗
17. Mangal, A. & Holm, E. A. A comparative study of feature selection methods for stress hotspot classification in materials. Integr. Mater. Manuf. Innov. 7, 87–95 (2018).10.1007/s40192-018-0109-8 ↗
18. Pion-Tonachini, L. et al. Learning from learning machines: a new generation of ai technology to meet the needs of science. Preprint at https://arxiv.org/abs/2111.13786 (2021).
19. Roscher, R., Bohn, B., Duarte, M. F. & Garcke, J. Explainable machine learning for scientific insights and discoveries. IEEE Access 8, 42200–42216 (2020).10.1109/ACCESS.2020.2976199 ↗
20. Raccuglia, P. et al. Machine-learning-assisted materials discovery using failed experiments. Nature 533, 73–76 (2016).10.1038/nature17439 ↗
21. Kondo, R., Yamakawa, S., Masuoka, Y., Tajima, S. & Asahi, R. Microstructure recognition using convolutional neural networks for prediction of ionic conductivity in ceramics. Acta Mater. 141, 29–38 (2017).10.1016/j.actamat.2017.09.004 ↗
22. Oviedo, F. et al. Fast and interpretable classification of small x-ray diffraction datasets using data augmentation and deep neural networks. npj Comput. Mater. 5, 60 (2019).10.1038/s41524-019-0196-x ↗
23. Oviedo, F., Ferres, J. L., Buonassisi, T. & Butler, K. T. Interpretable and explainable machine learning for materials science and chemistry. Acc. Mater. Res. 3, 597–607 (2022).10.1021/accountsmr.1c00244 ↗
24. Guidotti, R. et al. A survey of methods for explaining black box models. ACM Comput. Surv. 51, 1–42 (2018).10.1145/3236009 ↗
25. Arrieta, A. B. et al. Explainable artificial intelligence (xai): concepts, taxonomies, opportunities and challenges toward responsible ai. Inf. Fusion 58, 82–115 (2020).10.1016/j.inffus.2019.12.012 ↗
26. Kailkhura, B., Gallagher, B., Kim, S., Hiszpanski, A. & Han, T. Y.-J. Reliable and explainable machine-learning methods for accelerated material discovery. npj Comput. Mater. 5, 108 (2019).10.1038/s41524-019-0248-2 ↗
27. Pilania, G. Machine learning in materials science: From explainable predictions to autonomous design. Comput. Mater. Sci. 193, 110360 (2021).10.1016/j.commatsci.2021.110360 ↗
28. Lipton, Z. C. The mythos of model interpretability: in machine learning, the concept of interpretability is both important and slippery. Queue 16, 31–57 (2018).10.1145/3236386.3241340 ↗
29. Miller, T. Explanation in artificial intelligence: Insights from the social sciences. Artif. Intell. 267, 1–38 (2019).10.1016/j.artint.2018.07.007 ↗
30. Alvarez-Melis, D. & Jaakkola, T. S. Towards robust interpretability with self-explaining neural networks. In Advances in Neural Information Processing Systems (NIPS), vol. 31 (2018).
31. Gilpin, L. H. et al. Explaining explanations: an overview of interpretability of machine learning. In Proc. IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA), 80–89 (2018).10.1109/DSAA.2018.00018 ↗
32. Ouyang, R., Curtarolo, S., Ahmetcik, E., Scheffler, M. & Ghiringhelli, L. M. Sisso: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates. Phys. Rev. Mater. 2, 083802 (2018).10.1103/PhysRevMaterials.2.083802 ↗
33. Pankajakshan, P. et al. Machine learning and statistical analysis for materials science: stability and transferability of fingerprint descriptors and chemical insights. Chem. Mater. 29, 4190–4201 (2017).10.1021/acs.chemmater.6b04229 ↗
34. Gola, J. et al. Advanced microstructure classification by data mining methods. Comput. Mater. Sci. 148, 324–335 (2018).10.1016/j.commatsci.2018.03.004 ↗
35. Yang, C. et al. A machine learning-based alloy design system to facilitate the rational design of high entropy alloys with enhanced hardness. Acta Mater. 222, 117431 (2022).10.1016/j.actamat.2021.117431 ↗
36. Zhou, B., Khosla, A., Lapedriza, A., Oliva, A. & Torralba, A. Learning deep features for discriminative localization. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2921–2929 (2016).10.1109/CVPR.2016.319 ↗
37. Selvaraju, R. R. et al. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proc. IEEE International Conference on Computer Vision (ICCV), 618–626 (2017).10.1109/ICCV.2017.74 ↗
38. Lin, M., Chen, Q. & Yan, S. Network in network. In Proc. International Conference on Learning Representations (ICLR) (2014).
39. Gibson, I., Dransfield, G. & Irvine, J. Sinterability of commercial 8 molzirconia powders and the effect of sintered density on the ionic conductivity. J. Mater. Sci. 33, 4297–4305 (1998).10.1023/A:1004435504482 ↗
40. Chen, X., Khor, K., Chan, S. & Yu, L. Influence of microstructure on the ionic conductivity of yttria-stabilized zirconia electrolyte. Mater. Sci. Eng.: A 335, 246–252 (2002).10.1016/S0921-5093(01)01935-9 ↗
41. Ziletti, A., Kumar, D., Scheffler, M. & Ghiringhelli, L. M. Insightful classification of crystal structures using deep learning. Nat. Commun. 9, 1–10 (2018).10.1038/s41467-018-05169-6 ↗
42. Schütt, K. T., Arbabzadah, F., Chmiela, S., Müller, K. R. & Tkatchenko, A. Quantum-chemical insights from deep tensor neural networks. Nat. Commun. 8, 1–8 (2017).10.1038/ncomms13890 ↗
43. Hsu, T. et al. Efficient and interpretable graph network representation for angle-dependent properties applied to optical spectroscopy. npj Comput. Mater. 8, 151 (2022).10.1038/s41524-022-00841-4 ↗
44. Loveland, D., Liu, S., Kailkhura, B., Hiszpanski, A. & Han, Y. Reliable graph neural network explanations through adversarial training. In International Conference on Machine Learning (ICML) Workshop on Theoretic Foundation, Criticism, and Application Trend of Explainable AI (2021).
45. Pope, P. E., Kolouri, S., Rostami, M., Martin, C. E. & Hoffmann, H. Explainability methods for graph convolutional neural networks. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 10764–10773 (2019).10.1109/CVPR.2019.01103 ↗
46. Jung, Y.-J., Han, S.-H. & Choi, H.-J. Explaining cnn and rnn using selective layer-wise relevance propagation. IEEE Access 9, 18670–18681 (2021).10.1109/ACCESS.2021.3051171 ↗
47. Springenberg, J. T., Dosovitskiy, A., Brox, T. & Riedmiller, M. Striving for simplicity: the all convolutional net. In International Conference on Learning Representations (ICLR) Workshop (2015).
48. Sundararajan, M., Taly, A. & Yan, Q. Axiomatic attribution for deep networks. In Proc. International Conference on Machine Learning (ICML), 3319–3328 (2017).
49. Bach, S. et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE 10, e0130140 (2015).10.1371/journal.pone.0130140 ↗
50. Shrikumar, A., Greenside, P. & Kundaje, A. Learning important features through propagating activation differences. In Proc. International Conference on Machine Learning (ICML), 3145–3153 (2017).
51. Smilkov, D., Thorat, N., Kim, B., Viégas, F. & Wattenberg, M. Smoothgrad: removing noise by adding noise. In International Conference on Machine Learning (ICML) Workshop on Visualization for Deep Learning (2017).
52. Ancona, M., Ceolini, E., Öztireli, C. & Gross, M. Towards better understanding of gradient-based attribution methods for deep neural networks. In Proc. International Conference on Learning Representations (ICLR) (2018).10.1007/978-3-030-28954-6_9 ↗
53. Ribeiro, M. T., Singh, S. & Guestrin, C. “Why should i trust you?” explaining the predictions of any classifier. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135–1144 (2016).10.1145/2939672.2939778 ↗
54. Lundberg, S. M. & Lee, S.-I. A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (NIPS), vol. 30 (2017).
55. Gramegna, A. & Giudici, P. SHAP and LIME: an evaluation of discriminative power in credit risk. Front. Artif. Intell. 4, 752558 (2021).10.3389/frai.2021.752558 ↗
56. Man, X. & Chan, E. P. The best way to select features? comparing mda, lime, and shap. J. Financ. Data Sci. 3, 127–139 (2021).10.3905/jfds.2020.1.047 ↗
57. Iquebal, A. S., Pandagare, S. & Bukkapatnam, S. Learning acoustic emission signatures from a nanoindentation-based lithography process: towards rapid microstructure characterization. Tribol. Int. 143, 106074 (2020).10.1016/j.triboint.2019.106074 ↗
58. Giles, S. A., Sengupta, D., Broderick, S. R. & Rajan, K. Machine-learning-based intelligent framework for discovering refractory high-entropy alloys with improved high-temperature yield strength. Preprint at https://arxiv.org/abs/2112.02587 (2021).10.21203/rs.3.rs-1666028/v1 ↗
59. Newell, A. et al. Human Problem Solving, vol. 104 (Prentice-hall Englewood Cliffs, NJ, 1972).
60. Aamodt, A. & Plaza, E. Case-based reasoning: Foundational issues, methodological variations, and system approaches. AI Commun. 7, 39–59 (1994).10.3233/AIC-1994-7104 ↗
61. Zhong, X. et al. A study of real-world micrograph data quality and machine learning model robustness. npj Comput. Mater. 7, 161 (2021).10.1038/s41524-021-00616-3 ↗
62. Kim, B., Khanna, R. & Koyejo, O. O. Examples are not enough, learn to criticize! criticism for interpretability. In Advances in Neural Information Processing Systems (NIPS), vol. 29 (2016).
63. Quinlan, J. R. C4. 5: Programs for Machine Learning (Elsevier, 2014).
64. Hall, M. et al. The weka data mining software: an update. SIGKDD Explor., Newsl. 11, 10–18 (2009).10.1145/1656274.1656278 ↗
65. Picklum, M. & Beetz, M. Matcalo: knowledge-enabled machine learning in materials science. Comput. Mater. Sci. 163, 50–62 (2019).10.1016/j.commatsci.2019.03.005 ↗
66. Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).
67. Chen, T. & Guestrin, C. Xgboost: A scalable tree boosting system. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794 (2016).10.1145/2939672.2939785 ↗
68. Zilke, J. R., Mencía, E. L. & Janssen, F. Deepred – rule extraction from deep neural networks. In Proc. International Conference on Discovery Science, 457–473 (2016).10.1007/978-3-319-46307-0_29 ↗
69. Schmitz, G. P., Aldrich, C. & Gouws, F. S. Ann-dt: an algorithm for extraction of decision trees from artificial neural networks. IEEE Trans. Neural Netw. 10, 1392–1401 (1999).10.1109/72.809084 ↗
70. Augasta, M. G. & Kathirvalavakumar, T. Reverse engineering the neural networks for rule extraction in classification problems. Neural Process. Lett. 35, 131–150 (2012).10.1007/s11063-011-9207-8 ↗
71. Andrews, R., Diederich, J. & Tickle, A. B. Survey and critique of techniques for extracting rules from trained artificial neural networks. Knowl. -based Syst. 8, 373–389 (1995).10.1016/0950-7051(96)81920-4 ↗
72. Hailesilassie, T. Rule extraction algorithm for deep neural networks: a review. Int. J. Comput. Sci. Inf. Technol. 14, 376–381 (2016).
73. Ghiringhelli, L. M., Vybiral, J., Levchenko, S. V., Draxl, C. & Scheffler, M. Big data of materials science: critical role of the descriptor. Phys. Rev. Lett. 114, 105503 (2015).10.1103/PhysRevLett.114.105503 ↗
74. Pagan, D. C., Schmidt, G. H., Borum, A. D., Long, T. J. & Beaudoin, A. J. Informing mechanical model development using lower-dimensional descriptions of lattice distortion. Integr. Mater. Manuf. Innov. 9, 459–471 (2020).10.1007/s40192-020-00196-y ↗
75. Iwasaki, Y. et al. Machine-learning guided discovery of a new thermoelectric material. Sci. Rep. 9, 1–7 (2019).10.1038/s41598-019-39278-z ↗
76. Grigo, C. & Koutsourelakis, P.-S. A physics-aware, probabilistic machine learning framework for coarse-graining high-dimensional systems in the small data regime. J. Comput. Phys. 397, 108842 (2019).10.1016/j.jcp.2019.05.053 ↗
77. Iwasaki, Y. et al. Identification of advanced spin-driven thermoelectric materials via interpretable machine learning. npj Comput. Mater. 5, 103 (2019).10.1038/s41524-019-0241-9 ↗
78. Rovinelli, A., Sangid, M. D., Proudhon, H. & Ludwig, W. Using machine learning and a data-driven approach to identify the small fatigue crack driving force in polycrystalline materials. npj Comput. Mater. 4, 35 (2018).10.1038/s41524-018-0094-7 ↗
79. Schmidt, M. & Lipson, H. Distilling free-form natural laws from experimental data. Science 324, 81–85 (2009).10.1126/science.1165893 ↗
80. Blumer, A., Ehrenfeucht, A., Haussler, D. & Warmuth, M. K. Occam’s razor. Inf. Process. Lett. 24, 377–380 (1987).10.1016/0020-0190(87)90114-1 ↗
81. Pei, Z., Yin, J., Hawk, J. A., Alman, D. E. & Gao, M. C. Machine-learning informed prediction of high-entropy solid solution formation: beyond the hume-rothery rules. npj Comput. Mater. 6, 50 (2020).10.1038/s41524-020-0308-7 ↗
82. Ouyang, R. Sisso. https://github.com/rouyang2017/SISSO.
83. Erhan, D., Bengio, Y., Courville, A. & Vincent, P. Visualizing higher-layer features of a deep network. Univ. Montr. 1341, 1 (2009).
84. Simonyan, K., Vedaldi, A. & Zisserman, A. Deep inside convolutional networks: Visualising image classification models and saliency maps. In Proc. International Conference on Learning Representations (ICLR) Workshop (2014).
85. Holm, E. A. et al. Overview: computer vision and machine learning for microstructural characterization and analysis. Metall. Mater. Trans. A1-15 (2020).
86. Cecen, A., Dai, H., Yabansu, Y. C., Kalidindi, S. R. & Song, L. Material structure-property linkages using three-dimensional convolutional neural networks. Acta Mater. 146, 76–84 (2018).10.1016/j.actamat.2017.11.053 ↗
87. Jiao, Y., Stillinger, F. & Torquato, S. Modeling heterogeneous materials via two-point correlation functions: basic principles. Phys. Rev. 76, 031110 (2007).
88. Fullwood, D. T., Niezgoda, S. R. & Kalidindi, S. R. Microstructure reconstructions from 2-point statistics using phase-recovery algorithms. Acta Mater. 56, 942–948 (2008).10.1016/j.actamat.2007.10.044 ↗
89. Jha, D. et al. Elemnet: deep learning the chemistry of materials from only elemental composition. Sci. Rep. 8, 1–13 (2018).10.1038/s41598-018-35934-y ↗
90. Wang, A. Y.-T., Kauwe, S. K., Murdock, R. J. & Sparks, T. D. Compositionally restricted attention-based network for materials property predictions. npj Comput. Mater 7, 77 (2021).10.1038/s41524-021-00545-1 ↗
91. Vaswani, A. et al. Attention is all you need. In Advances in Neural Information Processing Systems (NIPS), vol. 30 (2017).
92. Kim, Y., Choi, Y., Widemann, D. & Zohdi, T. Efficient nonlinear manifold reduced order model. In Conference on Neural Information Processing Systems (NIPS) Workshop on Machine Learning for Engineering Modeling, Simulation and Design (2020).10.2172/1669223 ↗
93. Kim, Y., Choi, Y., Widemann, D. & Zohdi, T. A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder. J. Comput. Phys. 451, 110841 (2022).10.1016/j.jcp.2021.110841 ↗
94. Kadeethum, T. et al. Non-intrusive reduced order modeling of natural convection in porous media using convolutional autoencoders: comparison with linear subspace techniques. Adv. Water Resour. 160, 104098 (2022).10.1016/j.advwatres.2021.104098 ↗
95. Ling, J. et al. Building data-driven models with microstructural images: Generalization and interpretability. Mater. Discov. 10, 19–28 (2017).10.1016/j.md.2018.03.002 ↗
96. Deng, J. et al. Imagenet: A large-scale hierarchical image database. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 248–255 (2009).10.1109/CVPR.2009.5206848 ↗
97. Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. In Proc. 3rd International Conference on Learning Representations (ICLR) (2015).
98. Nguyen, A., Clune, J., Bengio, Y., Dosovitskiy, A. & Yosinski, J. Plug & play generative networks: conditional iterative generation of images in latent space. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 4467–4477 (2017).10.1109/CVPR.2017.374 ↗
99. Mahendran, A. & Vedaldi, A. Visualizing deep convolutional neural networks using natural pre-images. Int. J. Comput. Vis. 120, 233–255 (2016).10.1007/s11263-016-0911-8 ↗
100. Yosinski, J., Clune, J., Nguyen, A., Fuchs, T. & Lipson, H. Understanding neural networks through deep visualization. In Deep Learning Workshop at the 32nd International Conference on Machine Learning (ICML) (2015).