2023, Volume 24, Issue 5
Engineering >> 2023, Volume 24, Issue 5 doi: 10.1016/j.eng.2022.06.027
Fifth Paradigm in Science: A Case Study of an Intelligence-Driven Material Design
a Science and Technology on Parallel and Distributed Processing Laboratory, National University of Defense Technology, Changsha 410073, China
b Laboratory of Software Engineering for Complex Systems, National University of Defense Technology, Changsha 410073, China
c National Supercomputing Center in Changsha, Changsha 410082, China
d College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, China
e Institute of Chemical Biology and Nanomedicine, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, China
f Department of Applied Physics, School of Physics and Electronics, Hunan University, Changsha 410082, China
g Department of Computer Science, State University of New York, New Paltz, NY 12561, USA
Next Previous
Abstract
Science is entering a new era—the fifth paradigm—that is being heralded as the main character of knowledge integrating into different fields to intelligence-driven work in the computational community based on the omnipresence of machine learning systems. Here, we vividly illuminate the nature of the fifth paradigm by a typical platform case specifically designed for catalytic materials constructed on the Tianhe-1 supercomputer system, aiming to promote the cultivation of the fifth paradigm in other fields. This fifth paradigm platform mainly encompasses automatic model construction (raw data extraction), automatic fingerprint construction (neural network feature selection), and repeated iterations concatenated by the interdisciplinary knowledge ("volcano plot"). Along with the dissection is the performance evaluation of the architecture implemented in iterations. Through the discussion, the intelligence-driven platform of the fifth paradigm can greatly simplify and improve the extremely cumbersome and challenging work in the research, and realize the mutual feedback between numerical calculations and machine learning by compensating for the lack of samples in machine learning and replacing some numerical calculations caused by insufficient computing resources to accelerate the exploration process. It remains a challenging of the synergy of interdisciplinary experts and the dramatic rise in demand for on-the-fly data in data-driven disciplines. We believe that a glimpse of the fifth paradigm platform can pave the way for its application in other fields.
Keywords
Catalytic materials ; Fifth paradigm ; Intelligence-driven ; Machine learning ; Synergy of interdisciplinary experts
Figures
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
References
[ 1 ] Barber B. Resistance by scientists to scientific discovery. Science 1961;134(3479):596‒602. link1
[ 2 ] Dampier WCD. A history of science, technology and philosophy in the eighteenth century. Nature 1939;143(3613):134‒5. link1
[ 3 ] Crombie AC. Scientific change: historical studies in the intellectual, social and technical conditions for scientific discovery and technical invention, from antiquity to the present. London: Heinemann; 1963. link1
[ 4 ] Bidney M, Piekielek N. Towards a new paradigm in map and spatial information librarianship. J Map Geogr Libr 2018;14(2‒3):67‒74.
[ 5 ] Li J, Huang W. Paradigm shift in science with tackling global challenges. Natl Sci Rev 2019;6(6):1091‒3. link1
[ 6 ] Tolle KM, Tansley DSW, Hey AJG. The fourth paradigm: data-intensive scientific discovery. Proc IEEE 2011;99(8):1334‒7. link1
[ 7 ] Mnih V, Kavukcuoglu K, Silver D, Rusu AA, Veness J, Bellemare MG, et al.Human-level control through deep reinforcement learning. Nature 2015;518(7540):529‒33. link1
[ 8 ] Bainbridge WS. The scientific research potential of virtual worlds. Science 2007;317(5837):472‒6. link1
[ 9 ] Zubarev DY, Pitera JW. Cognitive materials discovery and onset of the 5th discovery paradigm. In: Pyzer-Knapp EO, Laino T, editors. Machine learning in chemistry: data-driven algorithms, learning systems, and predictions. Washington, DC: American Chemical Society; 2019. p. 103‒20. link1
[10] Malitsky N, Castain R, Cowan M. Spark‒MPI: approaching the fifth paradigm of cognitive applications. 2018. arXiv:1806.01110.
[11] Woinaroschy A. A paradigm-based evolution of chemical engineering. Chin J Chem Eng 2016;24(5):553‒7. link1
[12] Si Y, Wu HY, Yang K, Lian JC, Huang T, Huang WQ, et al. High-throughput computational design for 2D van der Waals functional heterostructures:fragility of Anderson’s rule and beyond. Appl Phys Lett 2021;119(4):043102. link1
[13] Li B, Peng W, Zhang J, Lian JC, Huang T, Cheng N, et al. High-throughput one-photon excitation pathway in 0D/3D heterojunctions for visible-light driven hydrogen evolution. Adv Funct Mater 2021;31(18):2100816. link1
[14] Himanen L, Geurts A, Foster AS, Rinke P. Data-driven materials science: status,challenges, and perspectives. Adv Sci 2019;6(21):1900808. Corrected in: Adv Sci 2020;7(2):1903667. link1
[15] Hardian R, Liang ZW, Zhang XL, Szekely G. Artificial intelligence: the silver bullet for sustainable materials development. Green Chem 2020;22(21):7521‒8. link1
[16] Xu X, Ma WP, Yan B. An electrodeposited nano-porous and neural network-like Ln@HOF film for SO2 gas quantitative detection via fluorescent sensing and machine learning. J Mater Chem A 2021;9(46):26391‒400. link1
[17] Kumar S, Ignacz G, Szekely G. Synthesis of covalent organic frameworks using sustainable solvents and machine learning. Green Chem 2021;23(22):8932‒9. link1
[18] Ding WL, Lu YM, Peng XL, Dong H, Chi WJ, Yuan X, et al. Accelerating evaluation of the mobility of ionic liquid-modulated PEDOT flexible electronics using machine learning. J Mater Chem A 2021;9(45):25547‒57. link1
[19] Vandenberg P. The fourth industrial revolution. J Asia Pac Econ 2020;25(1):194‒6. link1
[20] Feng R, Zhang C, Gao MC, Pei Z, Zhang F, Chen Y, et al. High-throughput design of high-performance lightweight high-entropy alloys. Nat Commun 2021;12(1):4329. link1
[21] Dobbelaere MR, Plehiers PP, Van de Vijver R, Stevens CV, Van Geem KM.Machine learning in chemical engineering: strengths, weaknesses,opportunities, and threats. Engineering 2021;7(9):1201‒11. link1
[22] Zhou T, Song Z, Sundmacher K. Big data creates new opportunities for materials research: a review on methods and applications of machine learning for materials design. Engineering 2019;5(6):1017‒26. link1
[23] Chen S, Zhang S, Shang J, Chen B, Zheng N. Brain inspired cognitive model with attention for self-driving cars. 2017. arXiv:1702.05596. link1
[24] Xu Z. Principle analysis of computer vision and its application research. In: Proceedings of the 2018 7th International Conference on Advanced Materials and Computer Science; 2018 Dec 21‒22; Dalian, China. Ottawa: Clausius Scientific Press; 2018. p. 478‒82. link1
[25] Itaya K, Takahashi K, Nakamura M, Koizumi M, Arakawa N, Tomita M, et al.BriCA: a modular software platform for whole brain architecture. In: Hirose A,Ozawa S, Doya K, Ikeda K, Lee M, Liu D, editors. Neural information processing. Cham: Springer International Publishing; 2016. p. 334‒41. link1
[26] US Department of Energy. Synergistic challenges in data-intensive science and exascale computing. Summary report of the Advanced Scientific Computing Advisory Committee (ASCAC) Subcommittee. Washington, DC: US Department of Energy, Office of Science; 2013. link1
[27] Wang C, Yu F, Liu Y, Li X, Chen J, Thiyagalingam J, et al. Deploying the Big Data Science Center at the Shanghai Synchrotron Radiation Facility: the first superfacility platform in China. Mach Learn Sci Technol 2021;2(3):035003. link1
[28] Tran K, Ulissi ZW. Active learning across intermetallics to guide discovery of electrocatalysts for CO2 reduction and H2 evolution. Nat Catal 2018;1(9):696‒703. link1
[29] Kresse G, Furthmüller J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput Mater Sci 1996;6(1):15‒50. link1
[30] Zhong M, Tran K, Min Y, Wang C, Wang Z, Dinh CT, et al. Accelerated discovery of CO2 electrocatalysts using active machine learning. Nature 2020;581 (7807):178‒83. link1
[31] Back S, Yoon J, Tian N, Zhong W, Tran K, Ulissi ZW. Convolutional neural network of atomic surface structures to predict binding energies for high-throughput screening of catalysts. J Phys Chem Lett 2019;10(15): 4401‒8. link1
[32] Wigner E, Seitz F. On the constitution of metallic sodium. Phys Rev 1933;43(10):804‒10. link1
[33] Abild-Pedersen F, Greeley J, Studt F, Rossmeisl J, Munter TR, Moses PG, et al. Scaling properties of adsorption energies for hydrogen-containing molecules on transition-metal surfaces. Phys Rev Lett 2007;99(1):016105. link1
[34] Calle-Vallejo F, Martínez JI, García-Lastra JM, Rossmeisl J, Koper MTM. Physical and chemical nature of the scaling relations between adsorption energies of atoms on metal surfaces. Phys Rev Lett 2012;108(11):116103. link1
[35] Hohenberg P, Kohn W. Inhomogeneous electron gas. Phys Rev 1964;136(3B): B864‒71. link1
[36] Kohn W, Sham LJ. Self-consistent equations including exchange and correlation effects. Phys Rev 1965;140(4A):A1133‒8. link1
[37] Tran K, Neiswanger W, Yoon J, Zhang Q, Xing E, Ulissi ZW. Methods for comparing uncertainty quantifications for material property predictions. Mach Learn Sci Technol 2020;1(2):025006. link1
[38] Garrido Torres JA, Jennings PC, Hansen MH, Boes JR, Bligaard T. Low-scaling algorithm for nudged elastic band calculations using a surrogate machine learning model. Phys Rev Lett 2019;122(15):156001. link1
[39] Chen C, Ye W, Zuo Y, Zheng C, Ong SP. Graph networks as a universal machine learning framework for molecules and crystals. Chem Mater 2019;31(9):3564‒72. link1
[40] Xie T, Grossman JC. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties. Phys Rev Lett 2018;120(14):145301. link1
[41] Gardner JR, Pleiss G, Bindel D, Weinberger KQ, Wilson AG. In: GPyTorch: blackbox matrix‒matrix Gaussian process inference with GPU acceleration. In: Bengio S, Wallach HM, Larochelle H, Grauman K, Cesa-Bianchi N, editors. Proceedings of the 32nd International Conference on Neural Information Processing Systems; 2018 Dec 3‒8. Montréal, QC, Canada. Red Hook: Curran Associates Inc.; 2018. p.7587‒97.
[42] Ong SP, Richards WD, Jain A, Hautier G, Kocher M, Cholia S, et al. Python materials genomics (pymatgen): a robust, open-source python library for materials analysis. Comput Mater Sci 2013;68:314‒9. link1
[43] Hjorth Larsen A, Jørgen Mortensen J, Blomqvist J, Castelli IE, Christensen R, Dułak M, et al. The atomic simulation environment—a Python library for working with atoms. J Phys Condens Matter 2017;29(27):273002. link1
[44] Jain A, Ong SP, Chen W, Medasani B, Qu X, Kocher M, et al. FireWorks: a dynamic workflow system designed for high-throughput applications. Concurr Comp Pract E 2015;27(17):5037‒59. link1
[45] Jiao YQ, Li YJ, Li B, Song YG, inventors; Inc.Goertek, assignee. [MongoDB-based test data storage query method and system]. Chinese patent CN 105550333A. 2021 May 4. Chinese.
[46] Wang Y, Lu Y, Qiu C, Gao P, Wang J. Performance evaluation of a infiniband-based lustre parallel file system. Proc Environ Sci 2011;11(Pt A):316‒21. link1
[47] Yoo AB, Jette MA, Grondona M. SLURM: simple Linux utility for resource management. In: Feitelson D, Rudolph L, Schwiegelshohn U, editors. Job scheduling strategies for parallel processing. Berlin: Springer; 2003. p. 44‒60. link1
[48] Nørskov JK, Bligaard T, Logadottir A, Kitchin JR, Chen JG, Pandelov S, et al. Trends in the exchange current for hydrogen evolution. J Electrochem Soc 2005;152(3):J23‒6. link1
[49] Chanussot L, Das A, Goyal S, Lavril T, Shuaibi M, Riviere M, et al. Open catalyst 2020 (oc20) dataset and community challenges. ACS Catal 2021;11(10):6059‒72. link1
京公网安备 11010502051620号
