2018
DOI: 10.1063/1.5048290
|View full text |Cite
|
Sign up to set email alerts
|

Machine learning enhanced global optimization by clustering local environments to enable bundled atomic energies

Abstract: We show how to speed up global optimization of molecular structures using machine learning methods. To represent the molecular structures we introduce the auto-bag feature vector that combines: i) a local feature vector for each atom, ii) an unsupervised clustering of such feature vectors for many atoms across several structures, and iii) a count for a given structure of how many times each cluster is represented. During subsequent global optimization searches, accumulated structure-energy relations of relaxed… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
35
0

Year Published

2019
2019
2022
2022

Publication Types

Select...
3
3
1

Relationship

0
7

Authors

Journals

citations
Cited by 47 publications
(35 citation statements)
references
References 56 publications
0
35
0
Order By: Relevance
“…To overcome this barrier, one way is to do a screen with some cheap methods (like using xTB in this paper); another way is to incorporate data science methods, like machine learning, to improve the global optimization algorithm. 95,96 Anyway, the new algorithm has made ABCluster a more convenient and powerful tool for solving theoretical and experimental problems in catalysis, material, and energy chemistry, etc.…”
Section: Discussionmentioning
confidence: 99%
“…To overcome this barrier, one way is to do a screen with some cheap methods (like using xTB in this paper); another way is to incorporate data science methods, like machine learning, to improve the global optimization algorithm. 95,96 Anyway, the new algorithm has made ABCluster a more convenient and powerful tool for solving theoretical and experimental problems in catalysis, material, and energy chemistry, etc.…”
Section: Discussionmentioning
confidence: 99%
“…Recently, Bayesian optimization (BO) (Shahriari et al 2015;Frazier 2018) is becoming one of the most widely adopted strategies for global optimization of multi-extremal, and expensive-to-evaluate objective functions related to, e.g., sensor networks (Garnett et al 2010), drug design (Meldgaard et al 2018), time-series forecasting (Candelieri et al 2018a), inversion problems (Perdikaris and Karniadakis 2016; Galuzzi et al 2018), and robotics (Olofsson et al 2018).…”
Section: Introductionmentioning
confidence: 99%
“…54,55 Hammer and co-workers showed how to combine evolutionary algorithms with ML techniques to find minimum energy structures, which can potentially be used for the targeted design of molecules. [56][57][58] Thus, there are many new potentially helpful algorithms for the theoretical chemical community, and it is not clear yet which of these new approaches will become a standard in the future. The aim of this work is therefore a prescreening of suitable ML algorithms that can be useful in the context of describing PESs, and the production of a large database to be used in further benchmarks.…”
Section: Introductionmentioning
confidence: 99%