scholarly journals Resolving Transition Metal Chemical Space: Feature Selection for Machine Learning and Structure–Property Relationships

2017 ◽  
Vol 121 (46) ◽  
pp. 8939-8954 ◽  
Author(s):  
Jon Paul Janet ◽  
Heather J. Kulik
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Transition Metal ◽  
Chemical Space ◽  
Structure Property ◽  
Structure Property Relationships ◽  
Selection For

scholarly journals Operator Quantum Machine Learning: Navigating the Chemical Space of Response Properties

2019 ◽  
Vol 73 (12) ◽  
pp. 1028-1031 ◽  
Author(s):  
Anders S. Christensen ◽  
O. Anatole von Lilienfeld
Keyword(s):  
Machine Learning ◽  
Chemical Space ◽  
Structure Property ◽  
Operator Formalism ◽  
Response Properties ◽  
Quantum Properties ◽  
Structure Property Relationships ◽  
Quantum Machine Learning ◽  
Common Response ◽  
Data Efficiency

The identification and use of structure–property relationships lies at the heart of the chemical sciences. Quantum mechanics forms the basis for the unbiased virtual exploration of chemical compound space (CCS), imposing substantial compute needs if chemical accuracy is to be reached. In order to accelerate predictions of quantum properties without compromising accuracy, our lab has been developing quantum machine learning (QML) based models which can be applied throughout CCS. Here, we briefly explain, review, and discuss the recently introduced operator formalism which substantially improves the data efficiency for QML models of common response properties.

scholarly journals Application of all relevant feature selection for failure analysis of parameter-induced simulation crashes in climate models

2015 ◽  
Vol 8 (7) ◽  
pp. 5419-5435 ◽  
Author(s):  
W. Paja ◽  
M. Wrzesień ◽  
R. Niemiec ◽  
W. R. Rudnicki
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Climate Models ◽  
Original Study ◽  
Relative Importance ◽  
Relevant Feature ◽  
Machine Learning Methods ◽  
Selection For ◽  
Robust Prediction ◽  
Physical Components

Abstract. The climate models are extremely complex pieces of software. They reflect best knowledge on physical components of the climate, nevertheless, they contain several parameters, which are too weakly constrained by observations, and can potentially lead to a crash of simulation. Recently a study by Lucas et al. (2013) has shown that machine learning methods can be used for predicting which combinations of parameters can lead to crash of simulation, and hence which processes described by these parameters need refined analyses. In the current study we reanalyse the dataset used in this research using different methodology. We confirm the main conclusion of the original study concerning suitability of machine learning for prediction of crashes. We show, that only three of the eight parameters indicated in the original study as relevant for prediction of the crash are indeed strongly relevant, three other are relevant but redundant, and two are not relevant at all. We also show that the variance due to split of data between training and validation sets has large influence both on accuracy of predictions and relative importance of variables, hence only cross-validated approach can deliver robust prediction of performance and relevance of variables.

scholarly journals Classification of Platinum Nanoparticle Catalysts using Machine Learning

2020 ◽  
Author(s):  
Amanda J. Parker ◽  
George Opletal ◽  
Amanda Barnard
Keyword(s):  
Machine Learning ◽  
Hydrogen Oxidation ◽  
Pt Nanoparticles ◽  
Structure Property ◽  
Metallic Nanoparticle ◽  
Platinum Nanoparticle ◽  
Nanoparticle Catalysts ◽  
Structure Property Relationships ◽  
Multi Stage

Computer simulations and machine learning provide complementary ways of identifying structure/property relationships that are typically targeting toward predicting the ideal singular structure to maximise the performance on a given application. This can be inconsistent with experimental observations that measure the collective properties of entire samples of structures that contain distributions or mixture of structures, even when synthesized and processed with care. Metallic nanoparticle catalysts are an important example. In this study we have used a multi-stage machine learning workflow to identify the correct structure/property relationships of Pt nanoparticles relevant to oxygen reduction (ORR), hydrogen oxidation (HOR) and hydrogen evolution (HER) reactions. By including classification prior to regression we identified two distinct classes of nanoparticles, and subsequently generate the class-specific models based on experimentally relevant criteria that are consistent with observations. These multi-structure/multi-property relationships, predicting properties averaged over a large sample of structures, provide a more accessible way to transfer data-driven predictions into the lab.

Sign in / Sign up

Export Citation Format

Share Document