https://scholars.lib.ntu.edu.tw/handle/123456789/576812
標題: | Learning to Optimize Molecular Geometries Using Reinforcement Learning | 作者: | Ahuja K Green W.H Li Y.-P. YI-PEI LI |
關鍵字: | article; cheminformatics; geometry; reinforcement learning (machine learning); software | 公開日期: | 2021 | 卷: | 17 | 期: | 2 | 起(迄)頁: | 818-825 | 來源出版物: | Journal of Chemical Theory and Computation | 摘要: | Though quasi-Newton methods have been widely adopted in computational chemistry software for molecular geometry optimization, it is well known that these methods might not perform well for initial guess geometries far away from the local minima, where the quadratic approximation might be inaccurate. We propose a reinforcement learning approach to develop a model that produces a correction term for the quasi-Newton step calculated with the BFGS algorithm to improve the overall optimization performance. Our model is able to complete the optimization in about 30% fewer steps than pure BFGS for molecules starting from perturbed geometries. The new method has similar convergence to BFGS when complemented with a line search procedure, but it is much faster since it avoids the multiple gradient evaluations associated with line searches. ? 2021 American Chemical Society. All rights reserved. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85100258282&doi=10.1021%2facs.jctc.0c00971&partnerID=40&md5=4182ca91e9d80654a9ab53cbd4e21973 https://scholars.lib.ntu.edu.tw/handle/123456789/576812 |
ISSN: | 15499618 | DOI: | 10.1021/acs.jctc.0c00971 |
顯示於: | 化學工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。