Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels
Abstract
A majority of experimental disciplines face the challenge of exploring large and high-dimensional parameter spaces in search of new scientific discoveries. Materials science is no exception; the wide variety of synthesis, processing, and environmental conditions that influence material properties gives rise to particularly vast parameter spaces. Recent advances have led to an increase in the efficiency of materials discovery by increasingly automating the exploration processes. Methods for autonomous experimentation have become more sophisticated recently, allowing for multi-dimensional parameter spaces to be explored efficiently and with minimal human intervention, thereby liberating the scientists to focus on interpretations and big-picture decisions. Gaussian process regression (GPR) techniques have emerged as the method of choice for steering many classes of experiments. We have recently demonstrated the positive impact of GPR-driven decision-making algorithms on autonomously-steered experiments at a synchrotron beamline. However, due to the complexity of the experiments, GPR often cannot be used in its most basic form, but rather has to be tuned to account for the special requirements of the experiments. Two requirements seem to be of particular importance, namely inhomogeneous measurement noise (input-dependent or non-i.i.d.) and anisotropic kernel functions, which are the two concepts that we tackle in this paper. Our synthetic andmore »
- Authors:
- Publication Date:
- Research Org.:
- Brookhaven National Laboratory (BNL), Upton, NY (United States); Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Basic Energy Sciences (BES); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- OSTI Identifier:
- 1678746
- Alternate Identifier(s):
- OSTI ID: 1693394; OSTI ID: 1713301
- Report Number(s):
- BNL-219997-2020-JAAM
Journal ID: ISSN 2045-2322; 17663; PII: 74394
- Grant/Contract Number:
- AC02-05CH11231; SC0012704
- Resource Type:
- Published Article
- Journal Name:
- Scientific Reports
- Additional Journal Information:
- Journal Name: Scientific Reports Journal Volume: 10 Journal Issue: 1; Journal ID: ISSN 2045-2322
- Publisher:
- Nature Publishing Group
- Country of Publication:
- United Kingdom
- Language:
- English
- Subject:
- 36 MATERIALS SCIENCE
Citation Formats
Noack, Marcus M., Doerk, Gregory S., Li, Ruipeng, Streit, Jason K., Vaia, Richard A., Yager, Kevin G., and Fukuto, Masafumi. Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels. United Kingdom: N. p., 2020.
Web. doi:10.1038/s41598-020-74394-1.
Noack, Marcus M., Doerk, Gregory S., Li, Ruipeng, Streit, Jason K., Vaia, Richard A., Yager, Kevin G., & Fukuto, Masafumi. Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels. United Kingdom. https://doi.org/10.1038/s41598-020-74394-1
Noack, Marcus M., Doerk, Gregory S., Li, Ruipeng, Streit, Jason K., Vaia, Richard A., Yager, Kevin G., and Fukuto, Masafumi. Mon .
"Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels". United Kingdom. https://doi.org/10.1038/s41598-020-74394-1.
@article{osti_1678746,
title = {Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels},
author = {Noack, Marcus M. and Doerk, Gregory S. and Li, Ruipeng and Streit, Jason K. and Vaia, Richard A. and Yager, Kevin G. and Fukuto, Masafumi},
abstractNote = {A majority of experimental disciplines face the challenge of exploring large and high-dimensional parameter spaces in search of new scientific discoveries. Materials science is no exception; the wide variety of synthesis, processing, and environmental conditions that influence material properties gives rise to particularly vast parameter spaces. Recent advances have led to an increase in the efficiency of materials discovery by increasingly automating the exploration processes. Methods for autonomous experimentation have become more sophisticated recently, allowing for multi-dimensional parameter spaces to be explored efficiently and with minimal human intervention, thereby liberating the scientists to focus on interpretations and big-picture decisions. Gaussian process regression (GPR) techniques have emerged as the method of choice for steering many classes of experiments. We have recently demonstrated the positive impact of GPR-driven decision-making algorithms on autonomously-steered experiments at a synchrotron beamline. However, due to the complexity of the experiments, GPR often cannot be used in its most basic form, but rather has to be tuned to account for the special requirements of the experiments. Two requirements seem to be of particular importance, namely inhomogeneous measurement noise (input-dependent or non-i.i.d.) and anisotropic kernel functions, which are the two concepts that we tackle in this paper. Our synthetic and experimental tests demonstrate the importance of both concepts for experiments in materials science and the benefits that result from including them in the autonomous decision-making process.},
doi = {10.1038/s41598-020-74394-1},
journal = {Scientific Reports},
number = 1,
volume = 10,
place = {United Kingdom},
year = {Mon Oct 19 00:00:00 EDT 2020},
month = {Mon Oct 19 00:00:00 EDT 2020}
}
https://doi.org/10.1038/s41598-020-74394-1
Works referenced in this record:
Mapping LUCAS topsoil chemical properties at European scale using Gaussian process regression
journal, December 2019
- Ballabio, Cristiano; Lugato, Emanuele; Fernández-Ugalde, Oihane
- Geoderma, Vol. 355
Accelerating materials property predictions using machine learning
journal, September 2013
- Pilania, Ghanshyam; Wang, Chenchen; Jiang, Xun
- Scientific Reports, Vol. 3, Issue 1
Beyond native block copolymer morphologies
journal, January 2017
- Doerk, Gregory S.; Yager, Kevin G.
- Molecular Systems Design & Engineering, Vol. 2, Issue 5
Data Processing at the Linac Coherent Light Source
conference, November 2019
- Thayer, Jana; Weninger, Clemens; Yamajala, Seshu
- 2019 IEEE/ACM 1st Annual Workshop on Large-scale Experiment-in-the-Loop Computing (XLOOP)
A Kriging-Based Approach to Autonomous Experimentation with Applications to X-Ray Scattering
journal, August 2019
- Noack, Marcus M.; Yager, Kevin G.; Fukuto, Masafumi
- Scientific Reports, Vol. 9, Issue 1
Dynamic X-ray diffraction sampling for protein crystal positioning
journal, January 2017
- Scarborough, Nicole M.; Godaliyadda, G. M. Dilshan P.; Ye, Dong Hye
- Journal of Synchrotron Radiation, Vol. 24, Issue 1
The Design and Analysis of Computer Experiments
book, January 2003
- Santner, Thomas J.; Williams, Brian J.; Notz, William I.
- Springer Series in Statistics
A Supervised Learning Approach for Dynamic Sampling
journal, February 2016
- Godaliyadda, G. M. Dilshan; Ye, Dong Hye; Uchic, Michael D.
- Electronic Imaging, Vol. 2016, Issue 19
Formula for the Viscosity of a Glycerol−Water Mixture
journal, May 2008
- Cheng, Nian-Sheng
- Industrial & Engineering Chemistry Research, Vol. 47, Issue 9
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation
journal, July 2013
- Jain, Anubhav; Ong, Shyue Ping; Hautier, Geoffroy
- APL Materials, Vol. 1, Issue 1
Most likely heteroscedastic Gaussian process regression
conference, January 2007
- Kersting, Kristian; Plagemann, Christian; Pfaff, Patrick
- Proceedings of the 24th international conference on Machine learning - ICML '07
Computational Strategies to Increase Efficiency of Gaussian-Process-Driven Autonomous Experiments
conference, November 2019
- Noack, Marcus; Zwart, Petrus
- 2019 IEEE/ACM 1st Annual Workshop on Large-scale Experiment-in-the-Loop Computing (XLOOP)
Advances in Kriging-Based Autonomous X-Ray Scattering Experiments
journal, January 2020
- Noack, Marcus M.; Doerk, Gregory S.; Li, Ruipeng
- Scientific Reports, Vol. 10, Issue 1
Rapid ordering of block copolymer thin films
journal, August 2016
- Majewski, Pawel W.; Yager, Kevin G.
- Journal of Physics: Condensed Matter, Vol. 28, Issue 40
A Variational Inference-Based Heteroscedastic Gaussian Process Approach for Simulation Metamodeling
journal, February 2019
- Wang, Wenjing; Chen, Nan; Chen, Xi
- ACM Transactions on Modeling and Computer Simulation, Vol. 29, Issue 1
Adaptive Strategies for Materials Design using Uncertainties
journal, January 2016
- Balachandran, Prasanna V.; Xue, Dezhen; Theiler, James
- Scientific Reports, Vol. 6, Issue 1
Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model
journal, July 2018
- Cang, Ruijin; Li, Hechao; Yao, Hope
- Computational Materials Science, Vol. 150
The meniscus-guided deposition of semiconducting polymers
journal, February 2018
- Gu, Xiaodan; Shaw, Leo; Gu, Kevin
- Nature Communications, Vol. 9, Issue 1
Preparation of Ordered Monolayers of Polymer Grafted Nanoparticles: Impact of Architecture, Concentration, and Substrate Surface Energy
journal, February 2016
- Che, Justin; Park, Kyoungweon; Grabowski, Christopher A.
- Macromolecules, Vol. 49, Issue 5
SAXS of self-assembled oriented lamellar nanocomposite films: an advanced method of evaluation
journal, July 2004
- Ruland, Wilhelm; Smarsly, Bernd
- Journal of Applied Crystallography, Vol. 37, Issue 4
Molecular Orientation and Grafting Density in Semifluorinated Self-Assembled Monolayers of Mono-, Di-, and Trichloro Silanes on Silica Substrates
journal, November 2002
- Genzer, Jan; Efimenko, Kirill; Fischer, Daniel A.
- Langmuir, Vol. 18, Issue 24