Machine learning approaches for the prediction of materials properties
Source abstract
Abstract
We give here a brief overview of the use of machine learning (ML) in our field, for chemists and materials scientists with no experience with these techniques. We illustrate the workflow of ML for computational studies of materials, with a specific interest in the prediction of materials properties. We present concisely the fundamental ideas of ML, and for each stage of the workflow, we give examples of the possibilities and questions to be considered in implementing ML-based modeling.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · X-ray Diffraction in Crystallography · Computational Drug Discovery Methods
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Chimie ParisTech, PSL University, CNRS, Institut de Recherche de Chimie Paris , 75005 Paris, France
Funding
Grand Équipement National de Calcul Intensif · Agence Nationale de la Recherche
References (73)
Show source-provided references
1. Computing machinery and intelligence10.1093/mind/lix.236.433 ↗
2. Some studies in machine learning using the game of checkers10.1147/rd.33.0210 ↗
3. Hey · 2009
4. Goertzel · 200710.1007/978-3-540-68677-4 ↗
5. Machine learning for molecular and materials science10.1038/s41586-018-0337-2 ↗
6. Crystallography open database—An open-access collection of crystal structures10.1107/s0021889809016690 ↗
7. Virtual exploration of the chemical universe up to 11 atoms of C, N, O, F: Assembly of 26.4 million structures (110.9 million stereoisomers) and analysis for new ring systems, stereochemistry, physicochemical properties, compound classes, and drug discovery10.1021/ci600423u ↗
8. ZINC 15—Ligand discovery for everyone10.1021/acs.jcim.5b00559 ↗
9. Commentary: The materials project: A materials genome approach to accelerating materials innovation10.1063/1.4812323 ↗
10. The AFLOW standard for high-throughput materials science calculations10.1016/j.commatsci.2015.07.019 ↗
11. The open quantum materials database (OQMD): Assessing the accuracy of DFT formation energies10.1038/npjcompumats.2015.10 ↗
12. The Harvard clean energy project: Large-scale computational screening and design of organic photovoltaics on the world community grid10.1021/jz200866s ↗
13. TE design lab: A virtual laboratory for thermoelectric material design10.1016/j.commatsci.2015.11.006 ↗
14. Correcting density functional theory for accurate predictions of compound enthalpies of formation: Fitted elemental-phase reference energies10.1103/physrevb.85.115104 ↗
15. L. Talirz , S.Kumbhar, E.Passaro, A. V.Yakutovich, V.Granata, F.Gargiulo, M.Borelli, M.Uhrin, S. P.Huber, S.Zoupanos, C. S.Adorf, C. W.Andersen, O.Schütt, C. A.Pignedoli, D.Passerone, J.VandeVondele, T. C.Schulthess, B.Smit, G.Pizzi, and N.Marzari, “Materials cloud, a platform for open computational science,” arXiv:2003.12510 [cond-mat.mtrl-sci] (2020).10.1038/s41597-020-00637-5 ↗
16. Python materials genomics (pymatgen): A robust, open-source python library for materials analysis10.1016/j.commatsci.2012.10.028 ↗
17. AiiDA: Automated interactive infrastructure and database for computational science10.1016/j.commatsci.2015.09.013 ↗
18. Systematic exploration of the mechanical properties of 13 621 inorganic compounds10.1039/c9sc01682a ↗
19. Scikit-learn: Machine learning in Python
20. Fast and accurate modeling of molecular atomization energies with machine learning10.1103/physrevlett.108.058301 ↗
21. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules10.1021/ci00057a005 ↗
22. Machine learning predictions of molecular properties: Accurate many-body potentials and nonlocality in chemical space10.1021/acs.jpclett.5b00831 ↗
23. Prediction errors of molecular machine learning models lower than hybrid DFT error10.1021/acs.jctc.7b00577 ↗
24. Communication: Understanding molecular representations in machine learning: The role of uniqueness and target similarity10.1063/1.4964627 ↗
25. How to represent crystal structures for machine learning: Towards fast prediction of electronic properties10.1103/physrevb.89.205118 ↗
26. Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations10.1103/physrevb.96.024104 ↗
27. Universal fragment descriptors for predicting properties of inorganic crystals10.1038/ncomms15679 ↗
28. Big data of materials science: Critical role of the descriptor10.1103/physrevlett.114.105503 ↗
29. Unsupervised machine learning in atomistic simulations, between predictions and understanding10.1063/1.5091842 ↗
30. Data mining for materials: Computational experiments with AB compounds10.1103/physrevb.85.104104 ↗
31. Machine learning methods to predict density functional theory B3LYP energies of HOMO and LUMO orbitals10.1021/acs.jcim.6b00340 ↗
32. Toward effective utilization of methane: Machine learning prediction of adsorption energies on metal alloys10.1021/acs.jpcc.7b12670 ↗
33. Rapid estimation of activation energy in heterogeneous catalytic reactions via machine learning10.1002/jcc.25567 ↗
34. Predicting catalytic activity of nanoparticles by a DFT-aided machine-learning algorithm10.1021/acs.jpclett.7b02010 ↗
35. Materials cartography: Representing and mining materials space using structural and electronic fingerprints10.1021/cm503507h ↗
36. Auto-generated materials database of Curie and Néel temperatures via semi-supervised relationship extraction10.1038/sdata.2018.111 ↗
37. Semi-supervised machine-learning classification of materials synthesis procedures10.1038/s41524-019-0204-1 ↗
38. Semi-supervised learning approaches to class assignment in ambiguous microstructures10.1016/j.actamat.2020.01.046 ↗
39. A statistical learning framework for materials science: Application to elastic moduli of k-nary inorganic polycrystalline compounds10.1038/srep34256 ↗
40. Predicting the mechanical properties of zeolite frameworks by machine learning10.1021/acs.chemmater.7b02532 ↗
41. Systematic investigation of the mechanical properties of pure silica zeolites: Stiffness, anisotropy, and negative linear compressibility10.1039/c3cp51817e ↗
42. Speeding up discovery of auxetic zeolite frameworks by machine learning10.1021/acs.chemmater.0c00434 ↗
43. Identifying systematic DFT errors in catalytic reactions10.1039/c5cy01332a ↗
44. Finding density functionals with machine learning10.1103/physrevlett.108.253002 ↗
45. Bypassing the Kohn–Sham equations with machine learning10.1038/s41467-017-00839-3 ↗
46. Can exact conditions improve machine-learned density functionals?10.1063/1.5025668 ↗
47. ωB97M-V: A combinatorially optimized, range-separated hybrid, meta-GGA density functional with VV10 nonlocal correlation10.1063/1.4952647 ↗
48. Machine learning adaptive basis sets for efficient large scale density functional theory simulation10.1021/acs.jctc.8b00378 ↗
49. Machine learning force field parameters from ab initio data10.1021/acs.jctc.7b00521 ↗
50. Ab initio derived force fields for zeolitic imidazolate frameworks: MOF-FF for ZIFs10.1021/acs.jctc.8b01041 ↗
51. Machine learning of accurate energy-conserving molecular force fields10.1126/sciadv.1603015 ↗
52. Towards exact molecular dynamics simulations with machine-learned force fields10.1038/s41467-018-06169-2 ↗
53. Machine learning force fields: Construction, validation, and outlook10.1021/acs.jpcc.6b10908 ↗
54. Property prediction of crystalline solids from composition and crystal structure10.1002/aic.15251 ↗
55. Machine learning energies of 2 million elpasolite (ABC2D6) crystals10.1103/physrevlett.117.135502 ↗
56. Machine learning for heterogeneous catalyst design and discovery10.1002/aic.16198 ↗
57. Application of DFT-based machine learning for developing molecular electrode materials in Li-ion batteries10.1039/c8ra07112h ↗
58. Cation–π and anion–π interactions10.1002/wcms.14 ↗
59. Deep learning in chemistry10.1021/acs.jcim.9b00266 ↗
60. Supervised self-organizing maps in crystal property and structure prediction10.1021/cg060872y ↗
61. PyBrain
62. Machine-learning-augmented chemisorption model for CO2 electroreduction catalyst screening10.1021/acs.jpclett.5b01660 ↗
63. Neural-symbolic machine learning for retrosynthesis and reaction prediction10.1002/chem.201605499 ↗
64. Generation of crystal structures using known crystal structures as analogues10.1107/s2052520616006533 ↗
65. Automatic chemical design using a data-driven continuous representation of molecules10.1021/acscentsci.7b00572 ↗
66. Charting a course for chemistry10.1038/s41557-019-0236-7 ↗
67. Synthetic organic chemistry driven by artificial intelligence10.1038/s41570-019-0124-0 ↗
68. Universal chemical synthesis and discovery with ‘the chemputer’10.1016/j.trechm.2019.07.004 ↗
69. Organic synthesis in a modular robotic system driven by a chemical programming language10.1126/science.aav2211 ↗
70. High-throughput discovery of organic cages and catenanes using computational screening fused with robotic synthesis10.1038/s41467-018-05271-9 ↗
71. Artificial-intelligence-driven organic synthesis—En route towards autonomous synthesis?10.1002/anie.201911062 ↗
72. A robotic platform for flow synthesis of organic compounds informed by AI planning10.1126/science.aax1566 ↗
73. Controlling an organic synthesis robot with machine learning to search for new reactivity10.1038/s41586-018-0307-8 ↗