When not to use machine learning: A perspective on potential and limitations

Carbone, Matthew R.

doi:10.1557/s43577-022-00417-z

When not to use machine learning: A perspective on potential and limitations

Journal Article · Fri Oct 21 04:00:00 EDT 2022 · MRS Bulletin

DOI:https://doi.org/10.1557/s43577-022-00417-z· OSTI ID:1894415

Abstract

The unparalleled success of artificial intelligence (AI) in the technology sector has catalyzed an enormous amount of research in the scientific community. It has proven to be a powerful tool, but as with any rapidly developing field, the deluge of information can be overwhelming, confusing, and sometimes misleading. This can make it easy to become lost in the same hype cycles that have historically ended in the periods of scarce funding and depleted expectations known as AI winters. Furthermore, although the importance of innovative, high-risk research cannot be overstated, it is also imperative to understand the fundamental limits of available techniques, especially in young fields where the rules appear to be constantly rewritten and as the likelihood of application to high-stakes scenarios increases. In this article, we highlight the guiding principles of data-driven modeling, how these principles imbue models with almost magical predictive power, and how they also impose limitations on the scope of problems they can address. Particularly, understanding when not to use data-driven techniques, such as machine learning, is not something commonly explored, but is just as important as knowing how to apply the techniques properly. We hope that the discussion to follow provides researchers throughout the sciences with a better understanding of when said techniques are appropriate, the pitfalls to watch for, and most importantly, the confidence to leverage the power they can provide.

Graphical abstract

View Journal Article

Sponsoring Organization:: USDOE

Grant/Contract Number:: SC0012704

OSTI ID:: 1894415

Alternate ID(s):: OSTI ID: 1895069

Journal Information:: MRS Bulletin, Journal Name: MRS Bulletin Journal Issue: 9 Vol. 47; ISSN 0883-7694

Publisher:: Cambridge University Press (CUP)Copyright Statement

Country of Publication:: United States

Language:: English

References (43)

Artificial Chemist: An Autonomous Quantum Dot Synthesis Bot Epps, Robert W.; Bowen, Michael S.; Volk, Amanda A. Advanced Materials, Vol. 32, Issue 30 https://doi.org/10.1002/adma.202001626	journal	June 2020
Autonomous Nanocrystal Doping by Self‐Driving Fluidic Micro‐Processors Bateni, Fazel; Epps, Robert W.; Antami, Kameel Advanced Intelligent Systems, Vol. 4, Issue 5 https://doi.org/10.1002/aisy.202200017	journal	March 2022
Machine learning in materials science Wei, Jing; Chu, Xuan; Sun, Xiang‐Yu InfoMat, Vol. 1, Issue 3 https://doi.org/10.1002/inf2.12028	journal	August 2019
Machine learning methods in chemoinformatics: Machine learning methods in chemoinformatics Mitchell, John B. O. Wiley Interdisciplinary Reviews: Computational Molecular Science, Vol. 4, Issue 5 https://doi.org/10.1002/wcms.1183	journal	February 2014
A logical calculus of the ideas immanent in nervous activity McCulloch, Warren S.; Pitts, Walter The Bulletin of Mathematical Biophysics, Vol. 5, Issue 4 https://doi.org/10.1007/BF02478259	journal	December 1943
The changing science of machine learning Langley, Pat Machine Learning, Vol. 82, Issue 3 https://doi.org/10.1007/s10994-011-5242-y	journal	February 2011
AI and Its New Winter: from Myths to Realities Floridi, Luciano Philosophy & Technology, Vol. 33, Issue 1 https://doi.org/10.1007/s13347-020-00396-6	journal	February 2020
An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for TiO2 Artrith, Nongnuch; Urban, Alexander Computational Materials Science, Vol. 114 https://doi.org/10.1016/j.commatsci.2015.11.047	journal	March 2016
Neural networks and statistical techniques: A review of applications Paliwal, Mukta; Kumar, Usha A. Expert Systems with Applications, Vol. 36, Issue 1 https://doi.org/10.1016/j.eswa.2007.10.005	journal	January 2009
Machine Learning in Computer-Aided Synthesis Planning Coley, Connor W.; Green, William H.; Jensen, Klavs F. Accounts of Chemical Research, Vol. 51, Issue 5 https://doi.org/10.1021/acs.accounts.8b00087	journal	April 2018
Machine Learning for Materials Scientists: An Introductory Guide toward Best Practices Wang, Anthony Yu-Tung; Murdock, Ryan J.; Kauwe, Steven K. Chemistry of Materials, Vol. 32, Issue 12 https://doi.org/10.1021/acs.chemmater.0c01907	journal	May 2020
Four Generations of High-Dimensional Neural Network Potentials Behler, Jörg Chemical Reviews https://doi.org/10.1021/acs.chemrev.0c00868	journal	March 2021
Supervised Machine-Learning-Based Determination of Three-Dimensional Structure of Metallic Nanoparticles Timoshenko, Janis; Lu, Deyu; Lin, Yuewei The Journal of Physical Chemistry Letters, Vol. 8, Issue 20 https://doi.org/10.1021/acs.jpclett.7b02364	journal	October 2017
Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules Gómez-Bombarelli, Rafael; Wei, Jennifer N.; Duvenaud, David ACS Central Science, Vol. 4, Issue 2 https://doi.org/10.1021/acscentsci.7b00572	journal	January 2018
SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules Weininger, David Journal of Chemical Information and Modeling, Vol. 28, Issue 1 https://doi.org/10.1021/ci00057a005	journal	February 1988
Human-level control through deep reinforcement learning Mnih, Volodymyr; Kavukcuoglu, Koray; Silver, David Nature, Vol. 518, Issue 7540 https://doi.org/10.1038/nature14236	journal	February 2015
Double-slit photoelectron interference in strong-field ionization of the neon dimer Kunitski, Maksim; Eicke, Nicolas; Huber, Pia Nature Communications, Vol. 10, Issue 1 https://doi.org/10.1038/s41467-018-07882-8	journal	January 2019
Random forest machine learning models for interpretable X-ray absorption near-edge structure spectrum-property relationships Torrisi, Steven B.; Carbone, Matthew R.; Rohr, Brian A. npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-00376-6	journal	July 2020
Theoretical prediction of high melting temperature for a Mo–Ru–Ta–W HCP multiprincipal element alloy Hong, Qi-Jun; Schroers, Jan; Hofmann, Douglas npj Computational Materials, Vol. 7, Issue 1 https://doi.org/10.1038/s41524-020-00473-6	journal	January 2021
Best practices in machine learning for chemistry Artrith, Nongnuch; Butler, Keith T.; Coudert, François-Xavier Nature Chemistry, Vol. 13, Issue 6 https://doi.org/10.1038/s41557-021-00716-z	journal	May 2021
Machine learning for molecular and materials science Butler, Keith T.; Davies, Daniel W.; Cartwright, Hugh Nature, Vol. 559, Issue 7715 https://doi.org/10.1038/s41586-018-0337-2	journal	July 2018
Re-epithelialization and immune cell behaviour in an ex vivo human skin model Rakita, Ana; Nikolić, Nenad; Mildner, Michael Scientific Reports, Vol. 10, Issue 1 https://doi.org/10.1038/s41598-019-56847-4	journal	January 2020
Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities Noack, Marcus M.; Zwart, Petrus H.; Ushizima, Daniela M. Nature Reviews Physics, Vol. 3, Issue 10 https://doi.org/10.1038/s42254-021-00345-y	journal	July 2021
Machine learning enabling high-throughput and remote operations at large-scale user facilities Konstantinova, Tatiana; Maffettone, Phillip M.; Ravel, Bruce Digital Discovery, Vol. 1, Issue 4 https://doi.org/10.1039/D2DD00014H	journal	January 2022
Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations Behler, Jörg Physical Chemistry Chemical Physics, Vol. 13, Issue 40 https://doi.org/10.1039/c1cp21668f	journal	January 2011
Atom-centered symmetry functions for constructing high-dimensional neural network potentials Behler, Jörg The Journal of Chemical Physics, Vol. 134, Issue 7 https://doi.org/10.1063/1.3553717	journal	February 2011
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation Jain, Anubhav; Ong, Shyue Ping; Hautier, Geoffroy APL Materials, Vol. 1, Issue 1 https://doi.org/10.1063/1.4812323	journal	July 2013
Accurate, affordable, and generalizable machine learning simulations of transition metal x-ray absorption spectra using the XANESNET deep neural network Rankine, C. D.; Penfold, T. J. The Journal of Chemical Physics, Vol. 156, Issue 16 https://doi.org/10.1063/5.0087255	journal	April 2022
Graph networks for molecular design Mercado, Rocío; Rastemo, Tobias; Lindelöf, Edvard Machine Learning: Science and Technology, Vol. 2, Issue 2 https://doi.org/10.1088/2632-2153/abcf91	journal	March 2021
I.—Computing Machinery and Intelligence Turing, A. M. Mind, Vol. LIX, Issue 236 https://doi.org/10.1093/mind/LIX.236.433	journal	October 1950
Machine-Learning X-Ray Absorption Spectra to Quantitative Accuracy Carbone, Matthew R.; Topsakal, Mehmet; Lu, Deyu Physical Review Letters, Vol. 124, Issue 15 https://doi.org/10.1103/PhysRevLett.124.156401	journal	April 2020
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces Behler, Jörg; Parrinello, Michele Physical Review Letters, Vol. 98, Issue 14 https://doi.org/10.1103/PhysRevLett.98.146401	journal	April 2007
Complex modeling: a strategy and software program for combining multiple information sources to solve ill posed structure and nanostructure inverse problems Juhás, Pavol; Farrow, Christopher L.; Yang, Xiaohao Acta Crystallographica Section A Foundations and Advances, Vol. 71, Issue 6, p. 562-568 https://doi.org/10.1107/S2053273315014473	journal	September 2015
Using a machine learning approach to determine the space group of a structure from the atomic pair distribution function Liu, Chia-Hao; Tao, Yunzhe; Hsu, Daniel Acta Crystallographica Section A Foundations and Advances, Vol. 75, Issue 4 https://doi.org/10.1107/S2053273319005606	journal	June 2019
Hands-On Bayesian Neural Networks—A Tutorial for Deep Learning Users Jospin, Laurent Valentin; Laga, Hamid; Boussaid, Farid IEEE Computational Intelligence Magazine, Vol. 17, Issue 2 https://doi.org/10.1109/MCI.2022.3155327	journal	May 2022
Self-driving laboratory for accelerated discovery of thin-film materials MacLeod, B. P.; Parlane, F. G. L.; Morrissey, T. D. Science Advances, Vol. 6, Issue 20 https://doi.org/10.1126/sciadv.aaz8867	journal	May 2020
Machine learning: Trends, perspectives, and prospects Jordan, M. I.; Mitchell, T. M. Science, Vol. 349, Issue 6245 https://doi.org/10.1126/science.aaa8415	journal	July 2015
Inverse molecular design using machine learning: Generative models for matter engineering Sanchez-Lengeling, Benjamin; Aspuru-Guzik, Alán Science, Vol. 361, Issue 6400 https://doi.org/10.1126/science.aat2663	journal	July 2018
Opportunities and Challenges for Machine Learning in Materials Science Morgan, Dane; Jacobs, Ryan Annual Review of Materials Research, Vol. 50, Issue 1 https://doi.org/10.1146/annurev-matsci-070218-010015	journal	July 2020
A survey of transfer learning Weiss, Karl; Khoshgoftaar, Taghi M.; Wang, DingDing Journal of Big Data, Vol. 3, Issue 1 https://doi.org/10.1186/s40537-016-0043-6	journal	May 2016
The machine learning revolution in materials? Reyes, Kristofer G.; Maruyama, Benji MRS Bulletin, Vol. 44, Issue 7 https://doi.org/10.1557/mrs.2019.153	journal	July 2019
Artificial intelligence for materials discovery Gomes, Carla P.; Selman, Bart; Gregoire, John M. MRS Bulletin, Vol. 44, Issue 7 https://doi.org/10.1557/mrs.2019.158	journal	July 2019
Advances in De Novo Drug Design: From Conventional to Machine Learning Methods Mouchlis, Varnavas D.; Afantitis, Antreas; Serra, Angela International Journal of Molecular Sciences, Vol. 22, Issue 4 https://doi.org/10.3390/ijms22041676	journal	February 2021

Similar Records

Exploring the role of judgement and shared situation awareness when working with AI recommender systems

Journal Article · Fri Jul 26 00:00:00 EDT 2024 · Cognition, Technology & Work · OSTI ID:2406706

Trust in Artificial Intelligence: Meta-Analytic Findings

Journal Article · Fri May 28 00:00:00 EDT 2021 · Human Factors · OSTI ID:2425834

When not to use machine learning: A perspective on potential and limitations

Citation Formats

References (43)

Similar Records

Related Subjects