1
|
Villavicencio N, Groves MN. Tuning Reinforcement Learning Parameters for Cluster Selection to Enhance Evolutionary Algorithms. ACS ENGINEERING AU 2024; 4:381-393. [PMID: 39185391 PMCID: PMC11342372 DOI: 10.1021/acsengineeringau.3c00068] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 11/16/2023] [Revised: 03/29/2024] [Accepted: 04/01/2024] [Indexed: 08/27/2024]
Abstract
The ability to find optimal molecular structures with desired properties is a popular challenge, with applications in areas such as drug discovery. Genetic algorithms are a common approach to global minima molecular searches due to their ability to search large regions of the energy landscape and decrease computational time via parallelization. In order to decrease the amount of unstable intermediate structures being produced and increase the overall efficiency of an evolutionary algorithm, clustering was introduced in multiple instances. However, there is little literature detailing the effects of differentiating the selection frequencies between clusters. In order to find a balance between exploration and exploitation in our genetic algorithm, we propose a system of clustering the starting population and choosing clusters for an evolutionary algorithm run via a dynamic probability that is dependent on the fitness of molecules generated by each cluster. We define four parameters, MFavOvrAll-A, MFavClus-B, NoNewFavClus-C, and Select-D, that correspond to a reward for producing the best structure overall, a reward for producing the best structure in its own cluster, a penalty for not producing the best structure, and a penalty based on the selection ratio of the cluster, respectively. A reward increases the probability of a cluster's future selection, while a penalty decreases it. In order to optimize these four parameters, we used a Gaussian distribution to approximate the evolutionary algorithm performance of each cluster and performed a grid search for different parameter combinations. Results show parameter MFavOvrAll-A (rewarding clusters for producing the best structure overall) and parameter Select-D (appearance penalty) have a significantly larger effect than parameters MFavClus-B and NoNewFavClus-C. In order to produce the most successful models, a balance between MFavOvrAll-A and Select-D must be made that reflects the exploitation vs exploration trade-off often seen in reinforcement learning algorithms. Results show that our reinforcement-learning-based method for selecting clusters outperforms an unclustered evolutionary algorithm for quinoline-like structure searches.
Collapse
Affiliation(s)
- Nathan Villavicencio
- Department
of Mathematics, California State University
Fullerton, Fullerton, California 92834, United States
| | - Michael N. Groves
- Department
of Chemistry and Biochemistry, California
State University Fullerton, Fullerton, California 92834, United States
| |
Collapse
|
2
|
Lourenço MP, Hostaš J, Herrera LB, Calaminici P, Köster AM, Tchagang A, Salahub DR. GAMaterial-A genetic-algorithm software for material design and discovery. J Comput Chem 2023; 44:814-823. [PMID: 36444916 DOI: 10.1002/jcc.27043] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/11/2022] [Revised: 09/26/2022] [Accepted: 11/06/2022] [Indexed: 11/30/2022]
Abstract
Genetic algorithms (GAs) are stochastic global search methods inspired by biological evolution. They have been used extensively in chemistry and materials science coupled with theoretical methods, ranging from force-fields to high-throughput first-principles methods. The methodology allows an accurate and automated structural determination for molecules, atomic clusters, nanoparticles, and solid surfaces, fundamental to understanding chemical processes in catalysis and environmental sciences, for instance. In this work, we propose a new genetic algorithm software, GAMaterial, implemented in Python3.x, that performs global searches to elucidate the structures of atomic clusters, doped clusters or materials and atomic clusters on surfaces. For all these applications, it is possible to accelerate the GA search by using machine learning (ML), the ML@GA method, to build subsequent populations. Results for ML@GA applied for the dopant distributions in atomic clusters are presented. The GAMaterial software was applied for the automatic structural search for the Ti6 O12 cluster, doping Al in Si11 (4Al@Si11 ) and Na10 supported on graphene (Na10 @graphene), where DFTB calculations were used to sample the complex search surfaces with reasonably low computational cost. Finally, the global search by GA of the Mo8 C4 cluster was considered, where DFT calculations were made with the deMon2k code, which is interfaced with GAMaterial.
Collapse
Affiliation(s)
- Maicon Pierre Lourenço
- Departamento de Química e Física - Centro de Ciências Exatas, Naturais e da Saúde - CCENS - Universidade Federal do Espírito Santo, Espírito Santo, Brazil
| | - Jiří Hostaš
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, Calgary, Alberta, Canada
| | - Lizandra Barrios Herrera
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, Calgary, Alberta, Canada
| | | | | | - Alain Tchagang
- Digital Technologies Research Centre, National Research Council of Canada, Ottawa, Ontario, Canada
| | - Dennis R Salahub
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, Calgary, Alberta, Canada
| |
Collapse
|
3
|
Du Z, Zhao H, Jia G, Li X. Design, fabrication, and evaluation of a large-area hybrid solar simulator for remote sensing applications. OPTICS EXPRESS 2023; 31:6184-6202. [PMID: 36823881 DOI: 10.1364/oe.482003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/28/2022] [Accepted: 01/21/2023] [Indexed: 06/18/2023]
Abstract
Solar irradiance variations have a direct effect on the accuracy and repeatability of identifying spectral signatures in the remote sensing field experiments. Solar simulators have been deployed to allow for testing under controlled and reproducible laboratory conditions. However, it is difficult and expensive to make a large-area solar simulation with the appropriate spectral content and spatial uniformity of irradiance. In this study, a hybrid solar simulator has been designed and constructed to provide large-area illumination for remote sensing simulation applications. A design method based on the two-phase genetic algorithm is proposed to improve the performance of the spectral match and spatial uniformity, which no longer relies on the traditional trial-and-error technique. The first phase is used to determine the most appropriate configuration of different lamps in order to represent the solar spectrum. The second phase is to accommodate an optimal placement of the multiple sources to achieve irradiance uniformity. Both numerical simulations and experiments were performed to verify the performances. The results showed that the solar simulator provided a good spectral match and spatial irradiance for simulating the variations in direct normal irradiance at different solar zenith angles. In addition, the modular design makes it possible to adjust irradiance on the target area without altering the spectral distribution. This work demonstrates the development and measurement of a hybrid solar simulator with a realizable optimal configuration of multiple lamps, and offers the prospect of a scalable, large-area solar simulation.
Collapse
|
4
|
Structure and Properties of 1237 Low-Lying Isomers of Magnesium Clusters Mgn (n = 2–32) Predicted with the DFT Global Optimization. J CLUST SCI 2022. [DOI: 10.1007/s10876-022-02291-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/18/2022]
|
5
|
Lourenço MP, Herrera LB, Hostaš J, Calaminici P, Köster AM, Tchagang A, Salahub DR. Automatic structural elucidation of vacancies in materials by active learning. Phys Chem Chem Phys 2022; 24:25227-25239. [DOI: 10.1039/d2cp02585j] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
Abstract
The artificial intelligence method based on active learning for the automatic structural elucidation of vacancies in materials. This is implemented in the quantum machine learning software/agent for material design and discovery (QMLMaterial).
Collapse
Affiliation(s)
- Maicon Pierre Lourenço
- Departamento de Química e Física – Centro de Ciências Exatas, Naturais e da Saúde – CCENS – Universidade Federal do Espírito Santo, 29500-000, Alegre, Espírito Santo, Brazil
| | - Lizandra Barrios Herrera
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, 2500 University Drive NW, Calgary, AB, T2N 1N4, Canada
| | - Jiří Hostaš
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, 2500 University Drive NW, Calgary, AB, T2N 1N4, Canada
| | - Patrizia Calaminici
- Departamento de Química, CINVESTAV, Av. Instituto Politécnico Nacional 2508, AP 14-740, México D.F. 07000, Mexico
| | - Andreas M. Köster
- Departamento de Química, CINVESTAV, Av. Instituto Politécnico Nacional 2508, AP 14-740, México D.F. 07000, Mexico
| | - Alain Tchagang
- Digital Technologies Research Centre, National Research Council of Canada, 1200 Montréal Road, Ottawa, ON, K1A 0R6, Canada
| | - Dennis R. Salahub
- Department of Chemistry, Department of Physics and Astronomy, CMS Centre for Molecular Simulation, IQST Institute for Quantum Science and Technology, Quantum Alberta, University of Calgary, 2500 University Drive NW, Calgary, AB, T2N 1N4, Canada
| |
Collapse
|
6
|
Ignatov SK, Belyaev SN, Panteleev SV, Masunov AE. How Many Isomers Do Metallic Clusters Have? Case of Magnesium Clusters of up to 55 Atoms. J Phys Chem A 2021; 125:6543-6555. [PMID: 34297565 DOI: 10.1021/acs.jpca.1c02529] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
Abstract
About 9000 structures of magnesium clusters Mgn (n = 2-13) generated via different methods were optimized at the DFT levels in order to estimate the number of all possible stable structures that can exist for the given cluster size (∼820,000 PES points were explored in total). It was found that the number of possible cluster isomers N quickly grows with a number of atoms n; however, it is significantly lower than the number of possible nonisomorphic graph structures, which can be drawn for the given n. At the DFT potential energy surface, we found only 543 local minima corresponding to the isomers of Mg2-Mg13. The number of isomers obtained in the DFT optimizations grows with n approximately as n4, whereas the N values extrapolated to the infinite generation process grow as n8. The cluster geometries obtained from the global DFT optimization were then used to adjust two empirical potentials of Gupta type (GP) and modified Sutton-Chen type (SCG3) describing the interactions between the magnesium atoms. Using these potentials, the extensive sets of structures Mg2-Mg55 (up to 30,000 clusters for each n) were optimized to obtain the dependence of the cluster isomer count on n in the continuous range of n = 2-30 and for selected n up to n = 55. It was found that the SCG3 potential, which is closer to the DFT results, gives a number of possible isomers growing as approximately n8.9, whereas GP potential results in the n4.3 dependence.
Collapse
Affiliation(s)
- Stanislav K Ignatov
- Lobachevsky State University of Nizhny Novgorod, 23 Gagarin Avenue, Nizhny Novgorod 603950, Russia
| | - Sergey N Belyaev
- Lobachevsky State University of Nizhny Novgorod, 23 Gagarin Avenue, Nizhny Novgorod 603950, Russia
| | - Sergey V Panteleev
- Lobachevsky State University of Nizhny Novgorod, 23 Gagarin Avenue, Nizhny Novgorod 603950, Russia
| | - Artëm E Masunov
- NanoScience Technology Center, University of Central Florida, Orlando, Florida 32826, United States.,South Ural State University, Lenin pr. 76, Chelyabinsk 454080, Russia.,National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), Kashirskoye shosse 31, Moscow 115409, Russia
| |
Collapse
|
7
|
|
8
|
Sun J, Yi J, Cheng L. Directional Monte Carlo Lattice Search Algorithm for the Structure Search of Alumina Clusters (Al2O3)n (n=1~50). ACTA CHIMICA SINICA 2021. [DOI: 10.6023/a21050207] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
|
9
|
Mashwani WK, Hamdi A, Asif Jan M, Göktaş A, Khan F. Large-scale global optimization based on hybrid swarm intelligence algorithm. JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2020. [DOI: 10.3233/jifs-192162] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Affiliation(s)
- Wali Khan Mashwani
- Institute of Numerical Sciences, Kohat University of Science & Technology, KPK, Pakistan
| | - Abdelouahed Hamdi
- Department of Mathematics, Statistics & Physics College of Arts and Sciences, University of Qatar, Doha, Qatar
| | - Muhammad Asif Jan
- Institute of Numerical Sciences, Kohat University of Science & Technology, KPK, Pakistan
| | - Atila Göktaş
- Department of Statistics, Mugla Sitki Kocman University, Turkey
| | - Fouzia Khan
- Institute of Numerical Sciences, Kohat University of Science & Technology, KPK, Pakistan
| |
Collapse
|
10
|
Mashwani WK, Khan A, Göktaş A, Unvan YA, Yaniay O, Hamdi A. Hybrid differential evolutionary strawberry algorithm for real-parameter optimization problems. COMMUN STAT-THEOR M 2020. [DOI: 10.1080/03610926.2020.1783559] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/23/2022]
Affiliation(s)
- Wali Khan Mashwani
- Institute of Numerical Sciences, Kohat University of Science & Technology, Kohat, Pakistan
| | - Abdullah Khan
- Institute of Numerical Sciences, Kohat University of Science & Technology, Kohat, Pakistan
| | - Atila Göktaş
- Department of Statistics, Muğla Sıtkı Koçman University, Bodrum, Turkey
| | - Yuksel Akay Unvan
- Department of Banking and Finance, Ankara Yildirim Beyazit University, Ankara, Turkey
| | - Ozgur Yaniay
- Department of Statistics, Hacettepe University, Ankara, Turkey
| | - Abdelouahed Hamdi
- Department of Mathematics, Statistics and Physics, Qatar University, Doha, Qatar
| |
Collapse
|