Hostname: page-component-586b7cd67f-rdxmf Total loading time: 0 Render date: 2024-11-30T05:50:13.888Z Has data issue: false hasContentIssue false
  • English
  • Français

Active-learning and materials design: the example of high glass transition temperature polymers

Published online by Cambridge University Press:  13 June 2019

Chiho Kim Open the ORCID record for Chiho Kim [Opens in a new window] ,
Anand Chandrasekaran ,
Anurag Jha  and
Rampi Ramprasad
Chiho Kim
Affiliation:
School of Materials Science and Engineering, Georgia Institute of Technology, 771 Ferst Drive NW, Atlanta, GA 30332, USA
Anand Chandrasekaran
Affiliation:
School of Materials Science and Engineering, Georgia Institute of Technology, 771 Ferst Drive NW, Atlanta, GA 30332, USA
Anurag Jha
Affiliation:
School of Materials Science and Engineering, Georgia Institute of Technology, 771 Ferst Drive NW, Atlanta, GA 30332, USA
Rampi Ramprasad*
Affiliation:
School of Materials Science and Engineering, Georgia Institute of Technology, 771 Ferst Drive NW, Atlanta, GA 30332, USA
*
Address all correspondence to Rampi Ramprasad at [email protected]
Get access

Abstract

Machine-learning (ML) approaches have proven to be of great utility in modern materials innovation pipelines. Generally, ML models are trained on predetermined past data and then used to make predictions for new test cases. Active-learning, however, is a paradigm in which ML models can direct the learning process itself through providing dynamic suggestions/queries for the “next-best experiment.” In this work, the authors demonstrate how an active-learning framework can aid in the discovery of polymers possessing high glass transition temperatures (Tg). Starting from an initial small dataset of polymer Tg measurements, the authors use Gaussian process regression in conjunction with an active-learning framework to iteratively add Tg measurements of candidate polymers to the training dataset. The active-learning framework employs one of three decision making strategies (exploitation, exploration, or balanced exploitation/exploration) for selection of the “next-best experiment.” The active-learning workflow terminates once 10 polymers possessing a Tg greater than a certain threshold temperature are selected. The authors statistically benchmark the performance of the aforementioned three strategies (against a random selection approach) with respect to the discovery of high-Tg polymers for this particular demonstrative materials design challenge.

Type
Artificial Intelligence Research Letters
Information
MRS Communications , Volume 9 , Issue 3 , September 2019 , pp. 860 - 866
Copyright
Copyright © Materials Research Society 2019 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

†

Chiho Kim and Anand Chandrasekaran equally contributed to this work.

References

1.Mannodi-Kanakkithodi, A., Huan, T.D., and Ramprasad, R.: Mining materials design rules from data: the example of polymer dielectrics. Chem. Mater. 29, 9001–9010 (2017).10.1021/acs.chemmater.7b02027Google Scholar
2.Huan, T.D., Boggs, S., Teyssedre, G., Laurent, C., Cakmak, M., Kumar, S., and Ramprasad, R.: Advanced polymeric dielectrics for high energy density applications. Prog. Mater. Sci. 83, 236–269 (2016).10.1016/j.pmatsci.2016.05.001Google Scholar
3.Mannodi-Kanakkithodi, A., Pilania, G., and Ramprasad, R.: Critical assessment of regression-based machine learning methods for polymer dielectrics. Comput. Mater. Sci. 125, 123–135 (2016).10.1016/j.commatsci.2016.08.039Google Scholar
4.Huan, T.D., Mannodi-Kanakkithodi, A., Kim, C., Sharma, V., Pilania, G., and Ramprasad, R.: A polymer dataset for accelerated property prediction and design. Sci. Data 3, 160012 (2016).10.1038/sdata.2016.12Google Scholar
5.Mannodi-Kanakkithodi, A., Pilania, G., Ramprasad, R., Lookman, T., and Gubernatis, J.E.: Multi-objective optimization techniques to design the pareto front of organic dielectric polymers. Comput. Mater. Sci. 125, 92–99 (2016).10.1016/j.commatsci.2016.08.018Google Scholar
6.Mannodi-Kanakkithodi, A., Pilania, G., Huan, T.D., Lookman, T., and Ramprasad, R.: Machine learning strategy for accelerated design of polymer dielectrics. Sci. Rep. 6, 20952 (2016).10.1038/srep20952Google Scholar
7.Mannodi-Kanakkithodi, A., Chandrasekaran, A., Kim, C., Huan, T.D., Pilania, G., Botu, V., and Ramprasad, R.: Scoping the polymer genome: a roadmap for rational polymer dielectrics design and beyond. Mater. Today 21, 785–796 (2018).10.1016/j.mattod.2017.11.021Google Scholar
8.Mannodi-Kanakkithodi, A., Treich, G.M., Huan, T.D., Ma, R., Tefferi, M., Cao, Y., Sotzing, G.A., and Ramprasad, R.: Rational co-design of polymer dielectrics for energy storage. Adv. Mater. 28, 6277–6291 (2016).10.1002/adma.201600377Google Scholar
9.Sharma, V., Wang, C.C., Lorenzini, R.G., Ma, R., Zhu, Q., Sinkovits, D.W., Pilania, G., Oganov, A.R., Kumar, S., Sotzing, G.A., Boggs, S.A., and Ramprasad, R.: Rational design of all organic polymer dielectrics. Nat. Commun. 5, 4845 (2014).10.1038/ncomms5845Google Scholar
10.Das, D., Chandrasekaran, A., Venkatram, S., and Ramprasad, R.: Effect of crystallinity on Li adsorption in polyethylene oxide. Chem. Mater. 30, 8804–8810 (2018).10.1021/acs.chemmater.8b03434Google Scholar
11.Ong, S.P., Andreussi, O., Wu, Y., Marzari, N., and Ceder, G.: Electrochemical windows of room-temperature ionic liquids from molecular dynamics and density functional theory calculations. Chem. Mater. 23, 2979–2986 (2011).10.1021/cm200679yGoogle Scholar
12.Warmuth, M.K., Liao, J., Rätsch, G., Mathieson, M., Putta, S., and Lemmen, C.: Active learning with support vector machines in the drug discovery process. J. Chem. Inf. Comput. Sci. 43, 667–673 (2003). PMID: 12653536.10.1021/ci025620tGoogle Scholar
13.Shahriari, B., Swersky, K., Wang, Z., Adams, R.P., and De Freitas, N.: Taking the human out of the loop: a review of Bayesian optimization. Proc. IEEE 104, 148–175 (2016).10.1109/JPROC.2015.2494218Google Scholar
14.Rouet-Leduc, B., Hulbert, C., Barros, K., Lookman, T., and Humphreys, C.J.: Automatized convergence of optoelectronic simulations using active machine learning. Appl. Phys. Lett. 111, 043506 (2017).10.1063/1.4996233Google Scholar
15.Yuan, R., Liu, Z., Balachandran, P.V., Xue, D., Zhou, Y., Ding, X., Sun, J., Xue, D., and Lookman, T.: Accelerated discovery of large electrostrains in BaTiO3-based piezo-electrics using active learning. Adv. Mater. 30, 1702884 (2018).10.1002/adma.201702884Google Scholar
16.Mueller, T., Kusne, A.G., and Ramprasad, R.: Machine learning in materials science: recent progress and emerging applications. In Reviews in Computational Chemistry, edited by Parrill, A.L. and Lipkowitz, K.B. (John Wiley & Sons, Inc., New York, 29, 2016), pp. 186–273.Google Scholar
17.Ramprasad, R., Batra, R., Pilania, G., Mannodi-Kanakkithodi, A., and Kim, C.: Machine learning in materials informatics: recent applications and prospects. npj Comput. Mater. 3, 54 (2017).10.1038/s41524-017-0056-5Google Scholar
18.Audus, D.J. and de Pablo, J.J.: Polymer informatics: opportunities and challenges. ACS Macro Lett. 6, 1078–1082 (2017).10.1021/acsmacrolett.7b00228Google Scholar
19.Peerless, J.S., Milliken, N.J., Oweida, T.J., Manning, M.D., and Yingling, Y.G.: Adv. Theory Simul. 2, 1800129 (2018).10.1002/adts.201800129Google Scholar
20.Thrun, S.: Handbook of Brain Science and Neural Networks (MIT Press, Cambridge, 1995), pp. 381–384.Google Scholar
21.Brandup, J., Immergut, E.H., and Grulke, E.A.: Polymer Handbook, 4th ed. (John Wiley and Sons, New York, 1999).Google Scholar
22.Bicerano, J.: Prediction of Polymer Properties (Marcel Dekker, Inc., New York, USA, 2002).10.1201/9780203910115Google Scholar
23.Polymer Properties Database. http://polymerdatabase.com, (accessed April 10, 2019).Google Scholar
24.Rouet-Leduc, B., Barros, K., Lookman, T., and Humphreys, C.J.: Optimization of GaN LEDs and the reduction of efficiency droop using active machine learning. Sci. Rep. 6, 24862 (2016).10.1038/srep24862Google Scholar
25.Bassman, L., Rajak, P., Kalia, R.K., Nakano, A., Sha, F., Sun, J., Singh, D.J., Aykol, M., Huck, P., Persson, K., and Vashishta, P.: Active learning for accelerated design of layered materials. npj Comput. Mater. 4, 74 (2018).10.1038/s41524-018-0129-0Google Scholar
26.Weininger, D.: SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J. Chem. Inf. Model. 28, 31–36 (1988).10.1021/ci00057a005Google Scholar
27.Kim, C., Chandrasekaran, A., Huan, T.D., Das, D., and Ramprasad, R.: Polymer genome: a data-powered polymer informatics platform for property predictions. J. Phys. Chem. C 122, 17575–17585 (2018).10.1021/acs.jpcc.8b02913Google Scholar
28.Pankajakshan, P., Sanyal, S., de Noord, O.E., Bhattacharya, I., Bhattacharyya, A., and Waghmare, U.: 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.6b04229Google Scholar
29.Huan, T.D., Mannodi-Kanakkithodi, A., and Ramprasad, R.: Accelerated materials property predictions and design using motif-based fingerprints. Phys. Rev. B 92, 14106 (2015).10.1103/PhysRevB.92.014106Google Scholar
30.Labute, P.: J. Mol. Graph. Model. 18, 464–477 (2000).10.1016/S1093-3263(00)00068-1Google Scholar
31.Ertl, P., Rohde, B., and Selzer, P.: Fast calculation of molecular polar surface area as a sum of fragment-based contributions and its application to the prediction of drug transport properties. J. Med. Chem. 43, 3714–3717 (2000).10.1021/jm000942eGoogle Scholar
32.Prasanna, S. and Doerksen, R.: Topological polar surface area: a useful descriptor in 2D-QSAR. Curr. Med. Chem. 16, 21–41 (2009).10.2174/092986709787002817Google Scholar
33.Nguyen, K., Blum, L., van Deursen, R., and Reymond, J-L.: Classification of organic molecules by molecular quantum numbers. ChemMedChem 4, 1803–1805 (2009).10.1002/cmdc.200900317Google Scholar
34.RDKit: Open Source Toolkit for Cheminformatics. http://www.rdkit.org/ (accessed April 10, 2019).Google Scholar
35.Forrester, A. and Sóbester, A.K.A.: Engineering Design via Surrogate Modelling (John Wiley and Sons, Chichester, West Sussex, 2008).10.1002/9780470770801Google Scholar