Abstract
Kernel ridge regression (KRR) is one of the most popular methods of non-linear regression analysis in quantum chemistry. One of the main ingredients of KRR is the representation of the underlying physical system which mainly determines the performance of predicting quantum-mechanical properties based on KRR. Several such representations have been developed for both, solids and molecules; all of them with different advantages and limitations. These descriptors correspond to a similarity measure between two chemical compounds which is represented by the kernel. As recent approaches define the kernel directly from the underlying physical system, it is important to understand the properties of kernels and how these kernel properties can be used to improve the performance of machine learning models for quantum chemistry. After reviewing key representations of molecules, we provide an intuition on how the choice of the kernel affects the model. This is followed by a more practical guide of two complementary kernel methods, one for supervised and one for unsupervised learning, respectively. Finally, we present a way to gain an understanding about the model complexity by estimating the effective dimensionality induced by the data, the representation, and the kernel.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
C. Cortes, V. Vapnik, Mach. Learn. 20(3), 273 (1995)
V. Vapnik, S.E. Golowich, A.J. Smola, in Advances in Neural Information Processing Systems (1997), pp. 281–287
K.-R. Müller, S. Mika, G. Rätsch, K. Tsuda, B. Schölkopf, IEEE Trans. Neural Netw. 12(2), 181 (2001). https://doi.org/10.1109/72.914517
B. Schölkopf, A.J. Smola, Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond (MIT Press, Cambridge, 2002)
B. Huang, O.A. von Lilienfeld, J. Chem. Phys. 145(16), 161102 (2016). https://doi.org/10.1063/1.4964627
B. Schölkopf, S. Mika, C.J. Burges, P. Knirsch, K.-R. Müller, G. Rätsch, A.J. Smola, IEEE Trans. Neural Netw. 10(5), 1000 (1999)
P. Indyk, R. Motwani, in Proceedings of the Thirtieth Annual ACM Symposium on Theory of Computing (ACM, New York, 1998), pp. 604–613
J.H. Friedman, Data Min. Knowl. Disc. 1(1), 55 (1997)
J. Rust, J. Econ. Soc. 1997, 487–516 (1997)
K.T. Schütt, H. Glawe, F. Brockherde, A. Sanna, K.R. Müller, E.K.U. Gross, Phys. Rev. B 89, 205118 (2014). https://doi.org/10.1103/PhysRevB.89.205118
S. Chmiela, H.E. Sauceda, I. Poltavsky, K.-R. Müller, A. Tkatchenko, Comput. Phys. Commun. 240, 38 (2019). https://doi.org/10.1016/j.cpc.2019.02.007
F.A. Faber, L. Hutchison, B. Huang, J. Gilmer, S.S. Schoenholz, G.E. Dahl, O. Vinyals, S. Kearnes, P.F. Riley, O.A. von Lilienfeld, J. Chem. Theory Comput. 13(11), 5255 (2017). https://doi.org/10.1021/acs.jctc.7b00577
F.A. Faber, A.S. Christensen, B. Huang, O.A. von Lilienfeld, J. Chem. Phys. 148(24), 241717 (2018). https://doi.org/10.1063/1.5020710
K.T. Schütt, F. Arbabzadah, S. Chmiela, K.R. Müller, A. Tkatchenko, Nat. Commun. 8, 13890 (2017)
K.T. Schütt, P.-J. Kindermans, H.E.S. Felix, S. Chmiela, A. Tkatchenko, K.-R. Müller, in Advances in Neural Information Processing Systems (2017), pp. 991–1001
G. Montavon, W. Samek, K.-R. Müller, Digit. Signal Process. 73, 1 (2018)
S. Chmiela, Towards exact molecular dynamics simulations with invariant machine-learned models. Dissertation, Technische Universität Berlin (2019). https://doi.org/10.14279/depositonce-8635
M. Rupp, A. Tkatchenko, K.-R. Müller, O.A. von Lilienfeld, Phys. Rev. Lett. 108, 058301 (2012). https://doi.org/10.1103/PhysRevLett.108.058301
K. Hansen, G. Montavon, F. Biegler, S. Fazli, M. Rupp, M. Scheffler, O.A. von Lilienfeld, A. Tkatchenko, K.-R. Müller, J. Chem. Theory Comput. 9(8), 3404 (2013). https://doi.org/10.1021/ct400195d
K. Hansen, F. Biegler, R. Ramakrishnan, W. Pronobis, O.A. von Lilienfeld, K.-R. Müller, A. Tkatchenko, J. Phys. Chem. Lett. 6(12), 2326 (2015). https://doi.org/10.1021/acs.jpclett.5b00831
W. Pronobis, K.T. Schütt, A. Tkatchenko, K.-R. Müller, Eur. Phys. J. B 91(8), 178 (2018). https://doi.org/10.1140/epjb/e2018-90148-y
W. Pronobis, A. Tkatchenko, K.-R. Müller, J. Chem. Theory Comput. 14(6), 2991 (2018). https://doi.org/10.1021/acs.jctc.8b00110
B.E. Boser, I.M. Guyon, V.N. Vapnik, in Proceedings of the Fifth Annual Workshop on Computational Learning Theory (ACM, New York, 1992), pp. 144–152
K.-R. Müller, A. Smola, G. Rätsch, B. Schölkopf, J. Kohlmorgen, V. Vapnik, in Advances in Kernel Methods—Support Vector Learning, pp. 243–254 (1999)
M. James, F.A. Russell, Philos. Trans. R. Soc. Lond. A 209(441–458), 415 (1909). https://doi.org/10.1098/rsta.1909.0016
A.J. Smola, B. Schölkopf, K.-R. Müller, Neural Netw. 11(4), 637 (1998). https://doi.org/10.1016/S0893-6080(98)00032-X
A. Zien, G. Rätsch, S. Mika, B. Schölkopf, T. Lengauer, K.-R. Müller, Bioinformatics 16(9), 799 (2000)
A.P. Bartók, R. Kondor, G. Csányi, Phys. Rev. B 87, 184115 (2013). https://doi.org/10.1103/PhysRevB.87.184115
G. Montavon, K. Hansen, S. Fazli, M. Rupp, F. Biegler, A. Ziehe, A. Tkatchenko, A.V. Lilienfeld, K.-R. Müller, in Advances in Neural Information Processing Systems (2012), pp. 440–448
R. Ramakrishnan, O.A. von Lilienfeld, CHIMIA Int. J. Chem. 69(4), 182 (2015)
G. Ferré, T. Haut, K. Barros, J. Chem. Phys. 146(11), 114107 (2017)
S. Chmiela, A. Tkatchenko, H.E. Sauceda, I. Poltavsky, K.T. Schütt, K.-R. Müller, Sci. Adv. 3(5), e1603015 (2017)
D. Hu, Y. Xie, X. Li, L. Li, Z. Lan, J. Phys. Chem. Lett. 9(11), 2725 (2018). https://doi.org/10.1021/acs.jpclett.8b00684
C.K. Williams, C.E. Rasmussen, Gaussian Processes for Machine Learning, vol. 2 (MIT Press, Cambridge, 2006)
B. Schölkopf, A. Smola, K.-R. Müller, in International Conference on Artificial Neural Networks (Springer, Berlin, 1997), pp. 583–588
Z. Liu, D. Chen, H. Bensmail, Biomed Res. Int. 2005(2), 155 (2005)
D. Antoniou, S.D. Schwartz, J. Phys. Chem. B 115(10), 2465 (2011)
B. Schölkopf, A. Smola, K. Müller, Neural Comput. 10(5), 1299 (1998). https://doi.org/10.1162/089976698300017467
Y.M. Koyama, T.J. Kobayashi, S. Tomoda, H.R. Ueda, Phys. Rev. E 78(4), 046702 (2008)
X. Han, IEEE/ACM Trans. Comput. Biol. Bioinform. 7(3), 537 (2010)
A. Varnek, I.I. Baskin, Mol. Inf. 30(1), 20 (2011)
X. Deng, X. Tian, S. Chen, Chemom. Intell. Lab. Syst. 127, 195 (2013)
M.L. Braun, J.M. Buhmann, K.-R. Müller, J. Mach. Learn. Res. 9, 1875 (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Pronobis, W., Müller, KR. (2020). Kernel Methods for Quantum Chemistry. In: Schütt, K., Chmiela, S., von Lilienfeld, O., Tkatchenko, A., Tsuda, K., Müller, KR. (eds) Machine Learning Meets Quantum Physics. Lecture Notes in Physics, vol 968. Springer, Cham. https://doi.org/10.1007/978-3-030-40245-7_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-40245-7_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-40244-0
Online ISBN: 978-3-030-40245-7
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)