Abstract-This paper provides an introduction to support vector machines (SVMs), kernel Fisher discriminant analysis, and kernel principal component analysis (PCA), as examples for successful kernel-based learning methods. We first give a short background about Vapnik-Chervonenkis (VC) theory and kernel feature spaces and then proceed to kernel based learning in supervised and unsupervised scenarios including practical and algorithmic considerations. We illustrate the usefulness of kernel algorithms by finally discussing applications such as optical character recognition (OCR) and DNA analysis.
Compounds of low lattice thermal conductivity (LTC) are essential for seeking thermoelectric materials with high conversion efficiency. Some strategies have been used to decrease LTC. However, such trials have yielded successes only within a limited exploration space. Here we report the virtual screening of a library containing 54,779 compounds. Our strategy is to search the library through Bayesian optimization using for the initial data the LTC obtained from first-principles anharmonic lattice dynamics calculations for a set of 101 compounds. We discovered 221 materials with very low LTC. Two of them have even an electronic band gap < 1 eV, what makes them exceptional candidates for thermoelectric applications. In addition to those newly discovered thermoelectric materials, the present strategy is believed to be powerful for many other applications in which chemistry of materials are required to be optimized.Thermoelectric generators are essential for utilizing otherwise waste heat. Because of the technological importance, researchers have been seeking materials with high conversion efficiency for decades [1][2][3][4]. Compounds of low lattice thermal conductivity (LTC) are essential for this purpose. Different strategies have been used to decrease LTC. Recently, high throughput screening (HTS) of materials using materials database constructed by first principles calculations has been recognized as an efficient tool for accelerated materials discovery [5][6][7][8][9]. Thanks to the recent progress of computational power and techniques, a large set of first principles calculations can be performed with the accuracy comparable to experiments. This is a straightforward strategy when both of the following conditions are satisfied: 1) the target physical property can be accurately computed by first principles methods. 2) The exploration space is well defined and not too large to compute the target physical property exhaustively in the space.In order to evaluate LTC with the accuracy comparable to experimental data, however, we need to develop a method that is far beyond the ordinary density functional theory (DFT) calculations. Since we need to treat multiple interactions among phonons, or anharmonic lattice dynamics, the computational cost is many orders of magnitudes higher than the ordinary DFT calculations. Such expensive calculations are practically possible only for a small number of simple compounds. HTS of a large DFT database of LTC is not a realistic approach unless the exploration space is narrowly confined. In the year 2014, Carrete and coworkers concentrated their efforts to search low LTC materials within half-Heusler compounds [10]. They made HTS of wide variety of halfHeusler compounds by examination of thermodynamical stability via DFT results. Then LTC was estimated either by full first principles calculations or by a machinelearning algorithm for a selected small number of compounds. HTS of low LTC using a quasiharmonic Debye model was also reported in 2014 [11]. Efficient prediction of LTC throu...
Automatic design of organic materials requires black-box optimization in a vast chemical space. In conventional molecular design algorithms, a molecule is built as a combination of predetermined fragments. Recently, deep neural network models such as variational autoencoders and recurrent neural networks (RNNs) are shown to be effective in de novo design of molecules without any predetermined fragments. This paper presents a novel Python library ChemTS that explores the chemical space by combining Monte Carlo tree search and an RNN. In a benchmarking problem of optimizing the octanol-water partition coefficient and synthesizability, our algorithm showed superior efficiency in finding high-scoring molecules. ChemTS is available at https://github.com/tsudalab/ChemTS.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.