Abstract
Active learning refers to collections of algorithms of systematically constructing the training dataset. It is closely related to uncertainty estimation—we, generally, do not need to train our model on samples on which our prediction already has low uncertainty. This chapter reviews active learning algorithms in the context of molecular modeling and illustrates their applications on practical problems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Small because evaluation of \(\bar {f}(x_i)\) is expensive.
- 2.
Here we have implicitly assumed that the distribution of f(θ, x) has zero mean.
- 3.
To be precise, it could be mathematically proved that only a limited number configurations will be added to the training set if the configurations are sampled from a distribution with a compact support.
References
N. Artrith, J. Behler, High-dimensional neural network potentials for metal surfaces: a prototype study for copper. Phys. Rev. B 85(4), 045439 (2012)
A.P. Bartók, R. Kondor, G. Csányi, On representing chemical environments. Phys. Rev. B 87(18), 184115 (2013)
C.M. Bishop, Bayesian neural networks. J. Braz. Comput. Soc. 4(1), 61–68 (1997)
S. Chmiela, H.E. Sauceda, K.-R. Muller, A. Tkatchenko, Towards exact molecular dynamics simulations with machine-learned force fields. Nat. Commun. 9(1), 1–10 (2018)
G. Csányi, T. Albaret, M. Payne, A. De Vita, Learn on the fly: a hybrid classical and quantum-mechanical molecular dynamics simulation. Phys. Rev. Lett. 93(17), 175503 (2004)
A. De Vita, R. Car, A novel scheme for accurate MD simulations of large systems, in MRS Proceedings, vol. 491 (Cambridge University Press, Cambridge, 1997), p. 473
Y. Gal, Uncertainty in deep learning. PhD thesis, University of Cambridge, 2016
Y. Gal, Z. Ghahramani, Dropout as a Bayesian approximation: representing model uncertainty in deep learning, in International Conference on Machine Learning (2016), pp. 1050–1059
Y. Gal, R. Islam, Z. Ghahramani, Deep Bayesian active learning with image data. in Proceedings of the 34th International Conference on Machine Learning, vol. 70 (2017), pp. 1183–1192. www.JMLR.org
K. Gubaev, E.V. Podryabinkin, A.V. Shapeev, Machine learning of molecular properties: locality and active learning. J. Chem. Phys. 148(24), 241727 (2018)
K. Gubaev, E.V. Podryabinkin, G.L. Hart, A.V. Shapeev, Accelerating high-throughput searches for new alloys with active learning of interatomic potentials. Comput. Mater. Sci. 156, 148–156 (2019)
K. Hansen, F. Biegler, R. Ramakrishnan, W. Pronobis, O.A. Von Lilienfeld, K.-R. Muller, A. Tkatchenko, Machine learning predictions of molecular properties: accurate many-body potentials and nonlocality in chemical space. J. Phys. Chem. Lett. 6(12), 2326–2331 (2015)
G.L.W. Hart, L.J. Nelson, R.W. Forcade, Generating derivative structures at a fixed concentration. Comput. Mater. Sci. 59, 101–107 (2012)
G.E. Hinton, N. Srivastava, A. Krizhevsky, I. Sutskever, R.R. Salakhutdinov, Improving neural networks by preventing co-adaptation of feature detectors (2012). Preprint. arXiv:1207.0580
G. Huang, Y. Li, G. Pleiss, Z. Liu, J.E. Hopcroft, K.Q. Weinberger, Snapshot ensembles: Train 1, get M for free. Paper presented at the 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, 24–26 April 2017. Conference Track Proceedings, 2017. https://openreview.net
H. Huo, M. Rupp, Unified representation for machine learning of molecules and crystals for machine learning (2017). Preprint. arXiv:1704.06439
M. Kampffmeyer, A.-B. Salberg, R. Jenssen, Semantic segmentation of small objects and modeling of uncertainty in urban remote sensing images using deep convolutional neural networks, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (2016), pp. 1–9
I. Kononenko, Bayesian neural networks. Biol. Cybern. 61(5), 361–370 (1989)
Z. Li, J.R. Kermode, A. De Vita, Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces. Phys. Rev. Lett. 114, 096405 (2015)
Z. Lu, J. Bongard, Exploiting multiple classifier types with active learning, in Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation (ACM, New York, 2009), pp. 1905–1906
A.O. Lyakhov, A.R. Oganov, H.T. Stokes, Q. Zhu, New developments in evolutionary structure prediction algorithm USPEX. Comput. Phys. Commun. 184(4), 1172–1182 (2013)
A.G. de G. Matthews, J. Hron, M. Rowland, R.E. Turner, Z. Ghahra-mani, Gaussian process behaviour in wide deep neural networks. Paper presented at the 6th International Conference on Learning Representations, ICLR 2018, Vancouver, Canada, 30 April–3 May 2018. Conference Track Proceedings, 2018. https://openreview.net
M.J. Mehl, D. Hicks, C. Toher, O. Levy, R.M. Hanson, G.L.W. Hart, S. Curtarolo, The AFLOW library of crystallographic prototypes: part 1. Comput. Mater. Sci. 136:S1–S828 (2017)
D. Molchanov, A. Ashukha, D. Vetrov, Variational dropout sparsifies deep neural networks, in Proceedings of the 34th International Conference on Machine Learning, vol. 70 (2017), pp. 2498–2507. www.JMLR.org
K. Neklyudov, D. Molchanov, A. Ashukha, D.P. Vetrov, Structured Bayesian pruning via log-normal multiplicative noise, in Advances in Neural Information Processing Systems (2017), pp. 6775–6784
A. Oganov (ed.), Modern Methods of Crystal Structure Prediction (Wiley-VCH, Weinheim, 2010)
D.W. Opitz, J.W. Shavlik, Generating accurate and diverse members of a neural-network ensemble, in Advances in Neural Information Processing Systems (1996), pp. 535–541
E.V. Podryabinkin, A.V. Shapeev, Active learning of linearly parametrized interatomic potentials. Comput. Mater. Sci. 140, 171–180 (2017)
E. Podryabinkin, E. Tikhonov, A. Shapeev, A. Oganov, Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning. Phys. Rev. B 99(6), 064114 (2019)
R. Ramakrishnan, P.O. Dral, M. Rupp, O.A. Von Lilienfeld, Quantum chemistry structures and properties of 134 kilo molecules. Sci. Data 1, 140022 (2014)
C.E. Rasmussen, Gaussian processes in machine learning, in Advanced Lectures on Machine Learning (Springer, Berlin, 2004), pp. 63–71
M.D. Richard, R.P. Lippmann, Neural network classifiers estimate Bayesian a posteriori probabilities. Neural Comput. 3(4), 461–483 (1991)
K. Schütt, P.-J. Kindermans, H.E.S. Felix, S. Chmiela, A. Tkatchenko, K.-R. Müller, SchNet: a continuous-filter convolutional neural network for modeling quantum interactions, in Advances in Neural Information Processing Systems (2017), pp. 991–1001
B. Settles, Active learning. Synth. Lect. Artif. Intell. Mach. Learn. 6(1), 1–114 (2012)
A.V. Shapeev, Moment tensor potentials: a class of systematically improvable interatomic potentials. Multiscale Model. Simul. 14(3), 1153–1173 (2016)
J.S. Smith, B. Nebgen, N. Lubbers, O. Isayev, A.E. Roitberg, Less is more: sampling chemical space with active learning. J. Chem. Phys. 148(24), 241733 (2018)
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, R. Salakhutdinov, Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)
M. Teye, H. Azizpour, K. Smith, Bayesian uncertainty estimation for batch normalized deep networks, ed. by J.G. Dy, A. Krause, in Proceedings of the 35th International Conference on Machine Learning, ICML2018, Stockholm, Sweden, 10–15 July 2018, vol. 80. Proceedings of Machine Learning Research PMLR (2018), pp. 4914–4923
E. Tsymbalov, M. Panov, A. Shapeev, Dropout-based active learning for regression, in International Conference on Analysis of Images, Social Networks and Texts (Springer, Cham, 2018), pp. 247–258
E. Tsymbalov, S. Makarychev, A. Shapeev, M. Panov, Deeper connections between neural networks and Gaussian processes speed-up active learning, in Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. Main track (2019), pp. 3599–3605
Z.-H. Zhou, J. Wu, W. Tang, Ensembling neural networks: many could be better than all. Artif. Intell. 137(1–2), 239–263 (2002)
Acknowledgements
The work was supported by the Skoltech NGP Program No. 2016-7/NGP (a Skoltech-MIT joint project). The authors acknowledge the usage of the Skoltech CEST cluster (Magnus) from Prof. Shapeev’s group for obtaining the results presented in this work.
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
Shapeev, A., Gubaev, K., Tsymbalov, E., Podryabinkin, E. (2020). Active Learning and Uncertainty Estimation. In: SchĂĽtt, K., Chmiela, S., von Lilienfeld, O., Tkatchenko, A., Tsuda, K., MĂĽller, 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_15
Download citation
DOI: https://doi.org/10.1007/978-3-030-40245-7_15
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)