Active learning-driven quantitative synthesis–structure–property relations for improving performance and revealing active sites of nitrogen-doped carbon for the hydrogen evolution reaction

Abstract

While quantitative structure-properties relations (QSPRs) have been developed successfully in multiple fields, catalyst synthesis affects structure and in turn performance, making simple QSPRs inadequate. Furthermore, catalysts often have multiple active...

Prediction and Optimization of Ion Transport Characteristics in Nanoparticle-Based Electrolytes Using Convolutional Neural Networks

“…Specifically, machine learning approaches have been employed to develop surrogate models for quantitative prediction of macroscopic properties with binary composite microstructure as input, represented by either low-dimensional structural descriptors or as representative volume ele-ments. In the studies pertaining to the former, the hand-crafted features representing the microstructure are linked to an associated property using a regression-based model [24][25][26][27][28][29][30] or artificial neural network. 31,32 For the latter case, convolutional neural networks (CNNs) 33 are employed to extract important features from the digitized images of composite microstructures, generated by stochastic growth of grains [34][35][36][37][38] or by random assignment of numbers in the uniformly voxelized grids, [39][40][41] in order to predict the effective property of interest.…”

NEXTorch: A Design and Bayesian Optimization Toolkit for Chemical Sciences and Engineering

“…Active learning refers to the idea of a machine learning algorithm "learning" from data, proposing next experiments or calculations, and improving prediction accuracy with fewer training data or lower cost. 2 Bayesian Optimization (BO), an active learning framework, often used to tune hyperparameters in machine learning models, has seen a rise in its applications to various chemical science fields, including parameter tuning for density functional theory (DFT) calculations, 3 catalyst synthesis, 4,5 high throughput reactions, 6 and computational material discovery. [7][8][9] Its close variant, kriging, 10 originating in geostatistics, has also been widely applied in process engineering.…”

“…These features allow human-in-the-loop design where decision-making on generating future data is aided by domain knowledge. Utilizing these features, NEXTorch can assist not only chemical synthesis in laboratory experiments 4 but also the multiscale computational tasks from molecular-scale design, such as heterogeneous catalyst calculations 17 and homogeneous (ligand) catalyst discovery, to reactor-scale optimization, i.e., automatic reactor optimization with computational fluid dynamics (CFD). 28 Third, NEXTorch is modular, making it easy to extend to other frameworks.…”

“…denotes the input variables (parameters). Here, = , , , … ; each of is a vector = ( 01 , 21 , … , 31 ) 4 ; is the number of sample points, and is the dimensionality of the input. Each parameter is bounded, and the resulting multidimensional space is called the design space A, i.e., ∈ A ∈ ℝ : .…”

NEXTorch: A Design and Bayesian Optimization Toolkit for Chemical Sciences and Engineering

“…Active learning refers to the idea of a machine learning algorithm "learning" from data, proposing next experiments or calculations, and improving prediction accuracy with fewer training data or lower cost. 2 Bayesian Optimization (BO), an active learning framework, often used to tune hyperparameters in machine learning models, has seen a rise in its applications to various chemical science fields, including parameter tuning for density functional theory (DFT) calculations, 3 catalyst synthesis, 4,5 high throughput reactions, 6 and computational material discovery. [7][8][9] Its close variant, kriging, 10 originating in geostatistics, has also been widely applied in process engineering.…”