Abstract
The ML approaches discussed thus far focused on the develo** composition-property models or to use these models toward the inverse design of materials. However, understanding the nature of these black-box models are important so that an informed decision making process can be employed. To this extent, in this chapter, interpretable ML models are discussed. First, SHAP, a post-hoc model agnostic approach is employed to interpret the composition property model. SHAP provides insights into the features governing a property both in a qualitative and quantitative manner. Further, SHAP also provides the coupling or interaction between the input features for a given property. Finally, the use of support vector machines to interpret the structure–dynamics relationships in materials through a novel machine-learned metric, namely, “softness” is discussed. Altogether, the chapter outlines how interpretable ML can be used to gain insights into the black-box functions for materials response.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
S.S. Schoenholz, Combining machine learning and physics to understand glassy systems. J. Phys. Conf. Ser. 1036, 012021 (2018). https://doi.org/10.1088/1742-6596/1036/1/012021. [Online]
S.S. Schoenholz, E.D. Cubuk, E. Kaxiras, A.J. Liu, Relationship between local structure and relaxation in out-of-equilibrium glassy systems. Proc. Natl. Acad. Sci. 114(2), 263–267 (2017)
E.D. Cubuk, S.S. Schoenholz, E. Kaxiras, A.J. Liu, Structural properties of defects in glassy liquids. J. Phys. Chem. B 120(26), 6139–6146 (2016)
S.S. Schoenholz, E.D. Cubuk, D.M. Sussman, E. Kaxiras, A.J. Liu, A structural approach to relaxation in glassy liquids. Nat. Phys. 12(5), 469–471 (2016)
E.D. Cubuk, S.S. Schoenholz, J.M. Rieser, B.D. Malone, J. Rottler, D.J. Durian, E. Kaxiras, A.J. Liu, Identifying structural flow defects in disordered solids using machine-learning methods. Phys. Rev. Lett. 114(10), 108001 (2015)
E.D. Cubuk, R. Ivancic, S.S. Schoenholz, D. Strickland, A. Basu, Z. Davidson, J. Fontaine, J.L. Hor, Y.-R. Huang, Y. Jiang et al., Structure-property relationships from universal signatures of plasticity in disordered solids. Science 358(6366), 1033–1037 (2017)
M.L. Falk, J.S. Langer, Dynamics of viscoplastic deformation in amorphous solids. Phys. Rev. E 57(6), 7192–7205 (1998). https://doi.org/10.1103/PhysRevE.57.7192. [Online]. https://link.aps.org/doi/10.1103/PhysRevE.57.7192
A. Smessaert, J. Rottler, Distribution of local relaxation events in an aging three-dimensional glass: Spatiotemporal correlation and dynamical het- erogeneity. Phys. Rev. E 88(2), 022314 (2013). https://doi.org/10.1103/PhysRevE.88.022314. [Online]. https://link.aps.org/doi/10.1103/PhysRevE.88.022314
R. Candelier, A. Widmer-Cooper, J.K. Kummerfeld, O. Dauchot, G. Biroli, P. Harrowell, D.R. Reichman, Spatiotemporal hierarchy of relaxation events, dynamical heterogeneities, and structural reorganization in a supercooled liquid. Phys. Rev. Lett. 105(13), 135702 (2010). https://doi.org/10.1103/PhysRevLett.105.135702. [Online]. https://link.aps.org/doi/10.1103/PhysRevLett.105.135702
C. Molnar, Interpretable Machine Learning. Lulu.com (2020)
S.M. Lundberg, S.-I. Lee, A unified approach to interpreting model predictions, in Proceedings of the 31st International Conference on Neural Information Processing Systems, pp. 4768–4777 (2017)
K. Yang, X. Xu, B. Yang, B. Cook, H. Ramos, N.M.A. Krishnan, M.M. Smedskjaer, C. Hoover, M. Bauchy, Predicting the Young’s modulus of silicate glasses using high-throughput molecular dynamics simulations and machine learning. Sci. Rep. 9(1), 8739 (2019). ISSN: 2045-2322. https://doi.org/10.1038/s41598-019-45344-3. [Online]. https://www.nature.com/articles/s41598-019-45344-3. Accessed 07 May 2019
M. Zaki, Jayadeva, N.A. Krishnan, Extracting processing and testing parameters from materials science literature for improved property prediction of glasses. Chem. Eng. Process. Process. Intensif. 108607 (2021). ISSN: 0255-2701. https://doi.org/10.1016/j.cep.2021.108607. [Online]. https://www.sciencedirect.com/science/article/pii/S0255270121003020
R. Bhattoo, S. Bishnoi, M. Zaki, N.A. Krishnan, Understanding the compositional control on electrical, mechanical, optical, and physical properties of inorganic glasses with interpretable machine learning. Acta Mater. 242, 118439 (2023)
M. Zaki, V. Venugopal, R. Bhattoo, S. Bishnoi, S.K. Singh, A.R. Allu, Jayadeva, N.A. Krishnan, Interpreting the optical properties of oxide glasses with machine learning and shapely additive explanations. J. Am. Ceram. Soc. 105(6), 4046–4057 (2022)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2024 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Krishnan, N.M.A., Kodamana, H., Bhattoo, R. (2024). Interpretable ML for Materials. In: Machine Learning for Materials Discovery. Machine Intelligence for Materials Science. Springer, Cham. https://doi.org/10.1007/978-3-031-44622-1_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-44622-1_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44621-4
Online ISBN: 978-3-031-44622-1
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)