Abstract
Thanks to recent technological advances, hydrogeologists now have access to large amounts of data acquired in real time. Processing these data using traditional modelling tools is difficult and poses a number of challenges especially for tasks such as extracting useful features, uncertainty quantification or identifying links between variables. Artificial intelligence, and more specifically its subset ‘machine learning (ML)’, may represent a way of the future in hydrogeological research and applications. Unfortunately, several aspects of machine-learning methods hamper its adoption as a complementary tool for hydrogeologists, namely the black-box nature of most models, an often-limited generalization ability, a hypothetical convergence, and uncertain transferability. Recently, an entirely novel paradigm in the field of machine learning has been identified—theory-guided machine learning–in which the models integrate some specific theoretical knowledge, laws or principles of the field of study. This review article sets out to examine three theory-guided methods in their ability to overcome the limitations of machine learning for hydrogeological research and applications. These methods are, respectively, theory-guided constrained optimization (TGCO), theory-guided refinement of outputs (TGRO) and theory-guided architecture (TGA). The analyses led to the following conclusions: the opacity of ML models can be reduced by any of the three theory-guided ML methods; convergence and generalizability can be enhanced by TGCO, TGA, or a combination of at least two of the theory-guided ML methods; and no study conducted to date has made it possible to deduce the effectiveness of these methods on the transferability of ML models.
Résumé
Grâce aux récentes avancées technologiques, les hydrogéologues ont désormais accès à de grandes quantités de données acquises en temps réel. Le traitement de ces données à l’aide d’outils de modélisation traditionnels est difficile et pose un certain nombre de défis, en particulier pour des tâches telles que l’extraction de caractéristiques utiles, la quantification de l’incertitude ou l’identification des liens entre les variables. L’intelligence artificielle et plus particulièrement son sous-ensemble « l’apprentissage automatique (AA) » peut représenter une voie d’avenir dans la recherche et les applications hydrogéologiques. Malheureusement, plusieurs aspects des méthodes d’apprentissage automatique entravent son adoption en tant qu’outil complémentaire pour les hydrogéologues, à savoir la nature boîte noire de la plupart des modèles, une capacité de généralisation souvent limitée, une convergence hypothétique et une transférabilité incertaine. Récemment, un paradigme entièrement nouveau dans le domaine de l’apprentissage automatique a été identifié—l’apprentissage automatique guidé par la théorie—dans lequel les modèles intègrent des connaissances théoriques, des lois ou des principes spécifiques du domaine d’étude. Le présent article de synthèse se propose d’examiner trois méthodes guidées par la théorie sous l’angle de leur capacité à surmonter les limites de l’apprentissage automatique pour la recherche et les applications hydrogéologiques. Ces méthodes sont respectivement l’optimisation contrainte guidée par la théorie (OCGT), le réajustement des sorties de modèle guidé par la théorie (RSGT) et l’architecture guidée par la théorie (AGT). Les analyses conduisent aux conclusions suivantes: l’opacité des modèles d’AA peut être réduite par l’une des trois méthodes d’AA guidé par la théorie; la convergence et la généralisation peuvent être améliorées par l’OCGT, l’AGT ou une combinaison d’au moins deux de ces méthodes d’AA guidé par la théorie; et aucune étude menée à ce jour n’a permis de conclure à l’efficacité de ces méthodes sur la transférabilité des modèles d’AA.
Resumen
Debido a los recientes avances tecnológicos, los hidrogeólogos tienen actualmente acceso a una gran cantidad de datos obtenidos en tiempo real. El tratamiento de estos datos con las herramientas tradicionales de modelización es difícil y plantea una serie de desafíos, especialmente para tareas como la extracción de características útiles, la cuantificación de la incertidumbre o la identificación de vínculos entre variables. La inteligencia artificial, y más concretamente su subconjunto “machine learning (ML)”, puede representar una vía de futuro en la investigación y las aplicaciones hidrogeológicas. Lamentablemente, varios aspectos de los métodos de machine learning obstaculizan su adopción como herramienta complementaria para los hidrogeólogos, a saber, la naturaleza de caja negra de la mayoría de los modelos, una capacidad de generalización a menudo limitada, una convergencia hipotética y una transferencia incierta. Recientemente, se ha identificado un paradigma totalmente novedoso en el campo del machine learning—el aprendizaje automático guiado por la teoría—en el que los modelos integran algunos conocimientos teóricos específicos, leyes o principios del campo de estudio. Este artículo de revisión se propone examinar tres métodos guiados por la teoría en su capacidad de superar las limitaciones del aprendizaje automático para la investigación y las aplicaciones hidrogeológicas. Estos métodos son, respectivamente, la optimización restringida orientada por la teoría (TGCO), el refinamiento de resultados orientado por la teoría (TGRO) y la arquitectura orientada por la teoría (TGA). Los análisis condujeron a las siguientes conclusiones: la transparencia de los modelos ML puede reducirse con cualquiera de los tres métodos ML guiados por la teoría; la convergencia y la generalización pueden mejorarse con TGCO, TGA o una combinación de al menos dos de los métodos ML orientados por la teoría; y ningún estudio realizado hasta la fecha ha permitido deducir la eficacia de estos métodos en la transferencia de los modelos ML.
摘要
由于近年的技术进步,水文地质学家现在可以访问大量实时获取的数据。使用传统建模工具处理这些数据很困难,并带来了许多挑战,尤其是对于提取有用特征、不确定性量化或识别变量之间的联系等任务。人工智能,更具体地说是其代名词“机器学习 (ML)传,可能代表了水文地质研究和应用的一种未来方式。不幸的是,机器学习方法的几个方面阻碍了它作为水文地质学家的辅助工具的采用,即大多数模型的黑箱性质、通常有限的概化能力、假设收敛性和不确定的可转移性。最近,机器学习领域出现了一种全新的范式——具有理论的机器学习——其中模型整合了研究领域的一些特定理论知识、规律或原则。这篇综述论文旨在研究三种理论指导方法克服机器学习方法在水文地质研究和应用方面的局限性的能力。这些方法分别是理论指导的约束优化 (TGCO)、理论指导的输出细化 (TGRO) 和理论指导的架构 (TGA)。分析得出以下结论:三种理论指导的 ML 方法中的任何一种都可以降低 ML 模型的不透明度; TGCO、TGA 或至少两种理论指导的 ML 方法的组合可以增强收敛性和泛化性;迄今为止,还没有任何研究能够推断出这些方法对 ML 模型的可转移性的有效性。
Resumo
Graças aos recentes avanços tecnológicos, os hidrogeólogos têm agora acesso a grandes quantidades de dados adquiridos em tempo real. O processamento desses dados com ferramentas tradicionais de modelagem é difícil e impõe uma série de desafios, especialmente para tarefas como extração de características úteis, quantificação de incerteza ou identificação de relações entre variáveis. A inteligência artificial, e mais especificamente seu subconjunto “aprendizagem de máquina (AM)”, pode representar uma oportunidade para o futuro na pesquisa e aplicações hidrogeológicas. Infelizmente, vários aspectos dos métodos de aprendizagem de máquinas impedem sua adoção como ferramenta complementar para hidrogeólogos, a saber, a natureza caixa preta da maioria dos modelos, uma capacidade de generalização muitas vezes limitada, uma hipotética convergência e uma transferibilidade incerta. Recentemente, foi identificado um paradigma inteiramente novo no campo da aprendizagem de máquinas—aprendizagem de máquinas orientada pela teoria—no qual os modelos integram alguns conhecimentos teóricos específicos, leis ou princípios do campo de estudo. Este artigo de revisão se propõe a examinar três métodos teoricamente orientados em sua capacidade de superar as limitações da aprendizagem de máquinas para pesquisas e aplicações hidrogeológicas. Estes métodos são, respectivamente, otimização limitada por teorias (OLPT), refinamento teórico de saídas (RTS) e arquitetura orientada por teorias (AOT). As análises levaram às seguintes conclusões: a opacidade dos modelos de AM pode ser reduzida por qualquer um dos três métodos de AM orientados pela teoria; a convergência e a generalizabilidade podem ser melhoradas por OLPT, AOT ou uma combinação de pelo menos dois dos métodos de AM orientados pela teoria; e nenhum estudo realizado até o momento permitiu deduzir a eficácia desses métodos sobre a transferibilidade dos modelos de AM.
Similar content being viewed by others
References
Adamowski J, Chan HF (2011) A wavelet neural network conjunction model for groundwater level forecasting. J Hydrol 407:28–40. https://doi.org/10.1016/j.jhydrol.2011.06.013
Afshar A, Mariño MA, Ebtehaj M, Moosavi J (2007) Rule-based fuzzy system for assessing groundwater vulnerability. J Environ Eng 133:532–540. https://doi.org/10.1061/(ASCE)0733-9372(2007)133:5(532)
Arabameri A, Roy J, Saha S, Blaschke T, Ghorbanzadeh O, Tien Bui D (2019) Application of probabilistic and machine learning models for groundwater potentiality mapping in Damghan sedimentary plain, Iran. Remote Sens 11. https://doi.org/10.3390/rs11243015
Arnold DN (2015) Stability, consistency, and convergence of numerical discretizations, encyclopedia of applied and computational mathematics. In: Engquist B (ed) Encyclopedia of applied and computational mathematics. Springer, Heidelberg, Germany, pp 1358–1364. https://doi.org/10.1007/978-3-540-70529-1_407
Ayodele TO (2010) Types of machine learning algorithms. New adv Mach Learn 3:19–48. https://doi.org/10.5772/9385
Bahareh K, Husam AHA-N, Biswajeet P, Vahideh S, Alfian Abdul H, Naonori U, Seyed Amir N (2019) Optimized conditioning factors using machine learning techniques for groundwater potential mapping. Water. https://doi.org/10.3390/w11091909
Bakshi S, de Lange E, van der Graaf P, Danhof M, Peletier L (2016) Understanding the behavior of systems pharmacology models using mathematical analysis of differential equations: prolactin modeling as a case study. CPT Pharmacometrics Syst Pharmacol 5:339–351. https://doi.org/10.1002/psp4.12098
Barzegar R, Fijani E, Asghari Moghaddam A, Tziritis E (2017) Forecasting of groundwater level fluctuations using ensemble hybrid multi-wavelet neural network-based models. Sci Total Environ. https://doi.org/10.1016/j.scitotenv.2017.04.189
Brodeur ZP, Herman JD, Steinschneider S (2020) Bootstrap aggregation and cross-validation methods to reduce overfitting in reservoir control policy search. Water Resour Res 56:e2020WR027184. https://doi.org/10.1029/2020WR027184
Ch S, Mathur S (2012) Groundwater level forecasting using SVM-PSO. Int J Hydrol Sci Technol 2:202–218. https://doi.org/10.1504/IJHST.2012.047432
Chang FJ, Huang CW, Chang LC, Kao IF (2016) Prediction of monthly regional groundwater levels through hybrid soft-computing techniques. J Hydrol 541:965–976. https://doi.org/10.1016/j.jhydrol.2016.08.006
Chen C, He W, Zhou H, Xue Y, Zhu M (2020a) A comparative study among machine learning and numerical models for simulating groundwater dynamics in the Heihe River basin, northwestern China. Sci Rep 10:3904. https://doi.org/10.1038/s41598-020-60698-9
Chen Y, Huang D, Zhang D, Zeng J, Wang N, Zhang H, Yan J (2020b) Theory-guided hard constraint projection (HCP): a knowledge-based data-driven scientific machine learning method. arXiv preprint arxiv-201206148. https://arxiv.org/abs/2012.06148. Accessed September 2021
Chesnaux R, Santoni S, Garel E, Huneau F (2018) An analytical method for assessing recharge using groundwater travel time in Dupuit-Forchheimer aquifers. Groundwater 56:986–992. https://doi.org/10.1111/gwat.12794
Clark P, Niblett T (1989) The CN2 induction algorithm. Mach Learn 3:261–283. https://doi.org/10.1023/A:1022641700528
Daw A, Thomas RQ, Carey CC, Read JS, Appling AP, Karpatne A (2020) Physics-guided architecture (pga) of neural networks for quantifying uncertainty in lake temperature modeling. Proceedings of the 2020 SIAM International Conference on Data Mining, SDM20, Cincinatti, OH, May 2020, pp 532–540
Feng S, Huo Z, Kang S, Tang Z, Wang F (2011) Groundwater simulation using a numerical model under different water resources management scenarios in an arid region of China. Environ Earth Sci 62:961–971. https://doi.org/10.1007/s12665-010-0581-8
Gorgij AD, Moghaddam AA, Kisi O (2017) Groundwater budget forecasting, using hybrid wavelet-ANN-GP modelling: a case study of Azarshahr plain, East Azerbaijan, Iran. Hydrol Res 48:455–467. https://doi.org/10.2166/nh.2016.202
Guo H, Zhuang X, Liang D, Rabczuk T (2020) Stochastic groundwater flow analysis in heterogeneous aquifer with modified neural architecture search (NAS) based physics-informed neural networks using transfer learning, arXiv preprint arXiv:201012344
Hautier G, Fischer CC, Jain A, Mueller T, Ceder G (2010) Finding nature’s missing ternary oxide compounds using machine learning and density functional theory. Chem Mater 22:3762–3767. https://doi.org/10.1021/cm100795d
Huang X, Gao L, Crosbie RS, Zhang N, Fu G, Doble R (2019) Groundwater recharge prediction using linear regression, multi-layer perception network, and deep learning. Water 11:1879. https://doi.org/10.3390/w11091879
John B, Das S (2020) Identification of risk zone area of declining piezometric level in the urbanized regions around the city of Kolkata based on ground investigation and GIS techniques. Groundw Sustain Dev 11:100354. https://doi.org/10.1016/j.gsd.2020.100354
Kadeethum T, Jørgensen TM, Nick HM (2020) Physics-informed neural networks for solving inverse problems of nonlinear Biot’s equations: batch training. arXiv preprint arXiv:200509638. https://arxiv.org/abs/2005.09638. Accessed
Kahana A, Turkel E, Dekel S, Givoli D (2020) Obstacle segmentation based on the wave equation and deep learning. J Comput Phys 413:109458. https://doi.org/10.1016/j.jcp.2020.109458
Karimpouli S, Tahmasebi P (2020) Physics informed machine learning: seismic wave equation. Geosci Front 11:1993–2001. https://doi.org/10.1016/j.gsf.2020.07.007
Karpatne A, Atluri G, Faghmous JH, Steinbach M, Banerjee A, Ganguly A, Shekhar S, Samatova N, Kumar V (2017) Theory-guided data science: a new paradigm for scientific discovery from data. IEEE Trans Knowl Data Eng 29:2318–2331. https://doi.org/10.1109/TKDE.2017.2720168
Kavvas ES, Yang L, Monk JM, Heckmann D, Palsson BO (2020) A biochemically-interpretable machine learning classifier for microbial GWAS. Nat Commun 11:1–11. https://doi.org/10.1038/s41467-020-16310-9
Khalil A, Almasri MN, McKee M, Kaluarachchi JJ (2005) Applicability of statistical learning algorithms in groundwater quality modeling. Water Resour Res 41. https://doi.org/10.1029/2004WR003608
Khandelwal A, Mithal V, Kumar V (2015) Post classification label refinement using implicit ordering constraint among data instances. 2015 IEEE International Conference on Data Mining IEEE, Atlantic City, NJ, November 2015, pp 799–804
Ling J, Kurzawski A, Templeton J (2016) Reynolds averaged turbulence modelling using deep neural networks with embedded invariance. J Fluid Mech 807:155–166. https://doi.org/10.1017/jfm.2016.615[Opens
Liu J, Wang K, Ma S, Huang J (2013) Accounting for linkage disequilibrium in genome-wide association studies: a penalized regression method. Stat Interface 6:99–115. https://doi.org/10.4310/SII.2013.v6.n1.a10
Meng X, Li Z, Zhang D, Karniadakis GE (2020) PPINN: Parareal physics-informed neural network for time-dependent PDEs. Comput Methods Appl Mech Eng 370:113250. https://doi.org/10.1016/j.cma.2020.113250
Moghaddam DD, Rahmati O, Panahi M, Tiefenbacher J, Darabi H, Haghizadeh A, Haghighi AT, Nalivan OA, Tien Bui D (2020) The effect of sample size on different machine learning models for groundwater potential mapping in mountain bedrock aquifers. Catena 187. https://doi.org/10.1016/j.catena.2019.104421
Mohamed A, Dan L, Kai S, Mohamed M, Aldaw E, Elubid B (2019) Hydrochemical analysis and fuzzy logic method for evaluation of groundwater quality in the North Chengdu plain, China. Int J Environ Res Public Health 16:302. https://doi.org/10.3390/ijerph16030302
Muralidhar N, Bu J, Cao Z, He L, Ramakrishnan N, Tafti D, Karpatne A (2020) PhyNet: physics guided neural networks for particle drag force prediction. In: Assembly Proceedings of the 2020 SIAM Int Conf Data Mining, Cincinatti, OH, May 2020, pp 559–567
Naghibi SA, Ahmadi K, Daneshi A (2017) Application of support vector machine, random forest, and genetic algorithm optimized random forest models in groundwater potential mapping. Water Resour Manag 31:2761–2775. https://doi.org/10.1007/s11269-017-1660-3
Nguyen C, Hassner T, Seeger M, Archambeau C (2020a) LEEP: a new measure to evaluate transferability of learned representations. Paper presented at the International Conference on Machine Learning, Vienna, July 2020
Nguyen PT, Ha DH, Jaafari A, Nguyen HD, Van Phong T, Al-Ansari N, Prakash I, Le HV, Pham BT (2020b) Groundwater potential mapping combining artificial neural network and real AdaBoost ensemble technique: the DakNong Province case-study, Vietnam. Int J Environ Res Public Health 17:2473. https://doi.org/10.3390/ijerph17072473
Nguyen PT, Ha DH, Nguyen HD, Van Phong T, Trinh PT, Al-Ansari N, Le HV, Pham BT, Ho LS, Prakash I (2020c) Improvement of Credal decision trees using ensemble frameworks for groundwater potential modeling. Sustainability 12:2622. https://doi.org/10.3390/su12072622
Oberlack M (2002) Symmetries and invariant solutions of turbulent flows and their implications for turbulence modelling. In: Oberlack M, Busse FH (ed) Theories of turbulence. Springer, Heidelberg, Germany, pp 301–366
Park Y, Ligaray M, Kim YM, Kim JH, Cho KH, Sthiannopkao S (2016) Development of enhanced groundwater arsenic prediction model using machine learning approaches in southeast Asian countries. Desalin Water Treat 57:12227–12236. https://doi.org/10.1080/19443994.2015.1049411
Pazzani MJ, Brunk CA (1991) Detecting and correcting errors in rule-based expert systems: an integration of empirical and explanation-based learning. Knowl Acquis 3:157–173. https://doi.org/10.1016/1042-8143(91)90003-6
Piccione A, Berkery J, Sabbagh S, Andreopoulos Y (2020) Physics-guided machine learning approaches to predict the ideal stability properties of fusion plasmas. Nuclear Fusion 60. https://doi.org/10.1088/1741-4326/ab7597
Pradhan S, Kumar S, Kumar Y, Sharma HC (2019) Assessment of groundwater utilization status and prediction of water table depth using different heuristic models in an Indian interbasin. Soft Computing 23:10261–10285. https://doi.org/10.1007/s00500-018-3580-4
Raazia S, Dar AQ (2021) A numerical model of groundwater flow in Karewa-alluvium aquifers of NW Indian Himalayan region. Model Earth Syst Environ. https://doi.org/10.1007/s40808-021-01126-3
Raissi M, Perdikaris P, Karniadakis GE (2019) Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J Comput Phys 378:686–707. https://doi.org/10.1016/j.jcp.2018.10.045
Rudin C (2019) Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat MachIntell 1:206–215. https://doi.org/10.1038/s42256-019-0048-x
Sahoo S, Jha MK (2013) Groundwater-level prediction using multiple linear regression and artificial neural network techniques: a comparative assessment. Hydrogeol J 21. https://doi.org/10.1007/s10040-013-1029-5
Sahoo S, Russo T, Elliott J, Foster I (2017) Machine learning algorithms for modeling groundwater level changes in agricultural regions of the US. Water Resour Res 53:3878–3895. https://doi.org/10.1002/2016WR019933
Sajedi-Hosseini F, Malekian A, Choubin B, Rahmati O, Cipullo S, Coulon F, Pradhan B (2018) A novel machine learning-based approach for the risk assessment of nitrate groundwater contamination. Sci Total Environ 644:954–962. https://doi.org/10.1016/j.scitotenv.2018.07.054
Shaham U, Yamada Y, Negahban S (2018) Understanding adversarial training: increasing local stability of supervised models through robust optimization. Neurocomputing 307:195–204. https://doi.org/10.1016/j.neucom.2018.04.027
Shiri J, Kisi O, Yoon H, Lee K-K, Nazemi AH (2013) Predicting groundwater level fluctuations with meteorological effect implications: a comparative study among soft computing techniques. Comput Geosci 56:32–44. https://doi.org/10.1016/j.cageo.2013.01.007
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learning Res 15:1929–1958
Su Y-S, Ni C-F, Li W-C, Lee I-H, Lin C-P (2020) Applying deep learning algorithms to enhance simulations of large-scale groundwater flow in IoTs. Appl Soft Comput 92:106298. https://doi.org/10.1016/j.asoc.2020.106298
Sun AY (2018) Discovering state-parameter mappings in subsurface models using generative adversarial networks. Geophys Res Lett 45:11137–111146. https://doi.org/10.1029/2018GL080404
Sun Y, Wang X, Tang X (2014) Deep learning face representation from predicting 10,000 classes. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, June 2014, pp 1891–1898
Tahmasebi P, Kamrava S, Bai T, Sahimi M (2020) Machine learning in geo- and environmental sciences: from small to large scale. Adv Water Resour 142:103619. https://doi.org/10.1016/j.advwatres.2020.103619
Tapoglou E, Trichakis IC, Dokou Z, Nikolos IK, Karatzas GP (2014) Groundwater-level forecasting under climate change scenarios using an artificial neural network trained with particle swarm optimization. Hydrol Sci J/J Sci Hydrol 59:1225–1239. https://doi.org/10.1080/02626667.2013.838005
Tartakovsky AM, Marrero CO, Perdikaris P, Tartakovsky GD, Barajas-Solano D (2020) Physics-informed deep neural networks for learning parameters and constitutive relationships in subsurface flow problems. Water Resour Res 56:e2019WR026731. https://doi.org/10.1029/2019WR026731
Tayfur G, Nadiri AA, Moghaddam AA (2014) Supervised intelligent committee machine method for hydraulic conductivity estimation. Water Resour Manag 28:1173–1184. https://doi.org/10.1007/s11269-014-0553-y
Tutmez B, Hatipoglu Z, Kaymak U (2006) Modelling electrical conductivity of groundwater using an adaptive neuro-fuzzy inference system. Comput Geosci 32:421–433. https://doi.org/10.1016/j.cageo.2005.07.003
Udrescu S-M, Tegmark M (2020) AI Feynman: a physics-inspired method for symbolic regression. Sci Adv 6:eaay2631. https://doi.org/10.1126/sciadv.aay2631
Urolagin S, Kv P, NVS R (2012) Generalization capability of artificial neural network incorporated with pruning method. In: Chandrasekaran K, Balakrishnan N, Thilagam PS (eds) Advanced computing, networking and security. Springer, Heidelberg, Germany, pp 171–178
Wang B, Oldham C, Hipsey MR (2016) Comparison of machine learning techniques and variables for groundwater dissolved organic nitrogen prediction in an urban area. Proced Eng 154:1176–1184. https://doi.org/10.1016/j.proeng.2016.07.527
Wang N, Chang H, Zhang D (2021) Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network. Comput Methods Appl Mech Eng 373:113492. https://doi.org/10.1016/j.cma.2020.113492
Wang N, Zhang D, Chang H, Li H (2020a) Deep learning of subsurface flow via theory-guided neural network. J Hydrol 584:124700. https://doi.org/10.1016/j.jhydrol.2020.124700
Wang R, Walters R, Yu R (2020b) Incorporating symmetry into deep dynamics models for improved generalization. arXiv preprint arXiv:200203061. https://arxiv.org/abs/1312.6197. Accessed September 2021
Warde-Farley D, Goodfellow IJ, Courville A, Bengio Y (2013) An empirical analysis of dropout in piecewise linear networks. arXiv preprint arXiv:13126197. https://arxiv.org/abs/1312.6197. Accessed September 2021
Willard J, Jia X, Xu S, Steinbach M, Kumar V (2020) Integrating physics-based modeling with machine learning: a survey. arXiv preprint arXiv:200304919. https://arxiv.org/abs/2003.04919. Accessed September 2021
Xu R, Zhang D, Rong M, Wang N (2021) Weak form theory-guided neural network (TgNN-wf) for deep learning of subsurface single- and two-phase flow. J Comput Phys 436:110318. https://doi.org/10.1016/j.jcp.2021.110318
Ying X (2019) An overview of overfitting and its solutions. Paper presented at the Journal of Physics, Conference Series, vol 1423. https://iopscience.iop.org/year/1742-6596/Y2019. Accessed September 2021
Yip KY, Gerstein M (2009) Training set expansion: an approach to improving the reconstruction of biological networks from limited and uneven reliable interactions. Bioinformatics 25:243–250. https://doi.org/10.1093/bioinformatics/btn602
Zhang P (2010) Industrial control system simulation routines, chap 19. In: Zhang P (ed) Advanced industrial control technology. Elsevier, Amsterdam, pp 781-810
Zobeiry N, Humfeld KD (2021) A physics-informed machine learning approach for solving heat transfer equation in advanced manufacturing and engineering applications. Eng Appl Artif Intell 101:104232. https://doi.org/10.1016/j.engappai.2021.104232
Zobeiry N, Reiner J, Vaziri R (2020a) Theory-guided machine learning for damage characterization of composites. Compos Struct. https://doi.org/10.1016/j.compstruct.2020.112407
Zobeiry N, Stewart A, Poursartip A (2020b) Applications of machine learning for process modeling of composites. Paper presented at the SAMPE Virtual Conference, 2020. https://www.nasampe.org/page/2020VirtualSeries. Accessed September 2020
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Adombi, A.V.D., Chesnaux, R. & Boucher, MA. Review: Theory-guided machine learning applied to hydrogeology—state of the art, opportunities and future challenges. Hydrogeol J 29, 2671–2683 (2021). https://doi.org/10.1007/s10040-021-02403-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10040-021-02403-2