The surface roughness of the c-TiO2 layer help controls the perovskite grain size without any other parameter. The direct effect of perovskite grain size on PSC performance is clarified.
Machine learning (ML) is increasingly used in chemical physics and materials science. One major area of thrust is machine learning of properties of molecules and solid materials from descriptors of chemical composition and structure. Recently, kernel regression methods of various flavors—such as kernel ridge regression, Gaussian process regression, and support vector machine—have attracted attention in such applications. Kernel methods allow benefiting simultaneously from the advantages of linear regressions and the superior expressive power of nonlinear kernels. In many applications, kernel methods are used in high-dimensional feature spaces, where sampling with training data is bound to be sparse and where effects specific to high-dimensional spaces significantly affect the performance of the method. We review recent applications of kernel-based methods for the prediction of properties of molecules and materials from descriptors of chemical composition and structure and related purposes. We discuss methodological aspects including choices of kernels appropriate for different applications, effects of dimensionality, and ways to balance expressive power and reliability of the model in high-dimensional feature spaces and with sparse data. We also discuss kernel regression-based hybrid ML approaches.
7 December 2022 in Physical Chemistry Chemical Physics
Machine learning (ML) based methods and tools have now firmly established themselves in physical chemistry and in particular in theoretical and computational chemistry and in materials chemistry. The generality of popular ML techniques such as neural networks or kernel methods (Gaussian process and kernel ridge regression and their flavors) permitted their application to diverse problems from prediction of properties of functional materials (catalysts, solid state ionic conductors, etc.) from descriptors to the building of interatomic potentials (where ML is currently routinely used in applications) and electron density functionals. These ML techniques are assumed to have superior expressive power of nonlinear methods, and are often used “as is”, with concepts such as “non-parametric” or “deep learning“ used without a clear justification for their need or advantage over simpler and more robust alternatives. In this Perspective, we highlight some interrelations between popular ML techniques and traditional linear regressions and basis expansions and demonstrate that in certain regimes (such as a very high dimensionality) these approximations might collapse. We also discuss ways to recover the expressive power of a nonlinear approach and to help select hyperparameters with the help of high-dimensional model representation and to obtain elements of insight while preserving the generality of the method.
Machine learning (ML) has found increasing use in physical sciences, including research on energy conversion and storage technologies, in particular, so-called sustainable technologies. While often ML is used to directly optimize the parameters or phenomena of interest in the space of features, in this perspective, we focus on using ML to construct objects and methods that help in or enable the modeling of the underlying phenomena. We highlight the need for machine learning from very sparse and unevenly distributed numeric data in multidimensional spaces in these applications. After a brief introduction of some common regression-type machine learning techniques, we focus on more advanced ML techniques which use these known methods as building blocks of more complex schemes and thereby allow working with extremely sparse data and also allow generating insight. Specifically, we will highlight the utility of using representations with subdimensional functions by combining the high-dimensional model representation ansatz with machine learning methods such as neural networks or Gaussian process regressions in applications ranging from heterogeneous catalysis to nuclear energy.
3 October 2022 in Journal of Mathematical Chemistry
Kernel methods such as kernel ridge regression and Gaussian process regression with Matern-type kernels have been increasingly used, in particular, to fit potential energy surfaces (PES) and density functionals, and for materials informatics. When the dimensionality of the feature space is high, these methods are used with necessarily sparse data. In this regime, the optimal length parameter of a Matern-type kernel may become so large that the method effectively degenerates into a low-order polynomial regression and, therefore, loses any advantage over such regression. This is demonstrated theoretically as well as numerically in the examples of six- and fifteen-dimensional molecular PES using squared exponential and simple exponential kernels. The results shed additional light on the success of polynomial approximations such as PIP for medium-size molecules and on the importance of orders-of-coupling-based models for preserving the advantages of kernel methods with Matern-type kernels of on the use of physically motivated (reproducing) kernels.