Back to Search
Start Over
Benchmarking the Performance of Bayesian Optimization across Multiple Experimental Materials Science Domains
- Source :
- npj Computational Materials, Vol 7, Iss 1, Pp 1-10 (2021)
- Publication Year :
- 2021
-
Abstract
- In the field of machine learning (ML) for materials optimization, active learning algorithms, such as Bayesian Optimization (BO), have been leveraged for guiding autonomous and high-throughput experimentation systems. However, very few studies have evaluated the efficiency of BO as a general optimization algorithm across a broad range of experimental materials science domains. In this work, we evaluate the performance of BO algorithms with a collection of surrogate model and acquisition function pairs across five diverse experimental materials systems, namely carbon nanotube polymer blends, silver nanoparticles, lead-halide perovskites, as well as additively manufactured polymer structures and shapes. By defining acceleration and enhancement metrics for general materials optimization objectives, we find that for surrogate model selection, Gaussian Process (GP) with anisotropic kernels (automatic relevance detection, ARD) and Random Forests (RF) have comparable performance and both outperform the commonly used GP without ARD. We discuss the implicit distributional assumptions of RF and GP, and the benefits of using GP with anisotropic kernels in detail. We provide practical insights for experimentalists on surrogate model selection of BO during materials optimization campaigns.
- Subjects :
- FOS: Computer and information sciences
Computer Science - Machine Learning
Active learning (machine learning)
FOS: Physical sciences
Machine learning
computer.software_genre
Field (computer science)
Machine Learning (cs.LG)
QA76.75-76.765
symbols.namesake
Surrogate model
General Materials Science
Computer software
Materials of engineering and construction. Mechanics of materials
Gaussian process
Condensed Matter - Materials Science
business.industry
Bayesian optimization
Materials Science (cond-mat.mtrl-sci)
Function (mathematics)
Computer Science Applications
Random forest
Range (mathematics)
Mechanics of Materials
Modeling and Simulation
Physics - Data Analysis, Statistics and Probability
TA401-492
symbols
Artificial intelligence
business
computer
Data Analysis, Statistics and Probability (physics.data-an)
Subjects
Details
- Language :
- English
- Database :
- OpenAIRE
- Journal :
- npj Computational Materials, Vol 7, Iss 1, Pp 1-10 (2021)
- Accession number :
- edsair.doi.dedup.....5666907b95353455d90e05f53af6ea45