{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:48:10Z","timestamp":1768510090210,"version":"3.49.0"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T00:00:00Z","timestamp":1768435200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100014266","name":"Shell Brasil","doi-asserted-by":"publisher","award":["21373-6"],"award-info":[{"award-number":["21373-6"]}],"id":[{"id":"10.13039\/501100014266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Causal reasoning is essential for understanding relationships and guiding decision-making in different applications, as it allows for the identification of cause-and-effect relationships between variables. By uncovering the underlying process that drives these relationships, causal reasoning enables more accurate predictions, controlled interventions, and the ability to distinguish genuine causal effects from mere correlations in complex systems. In oil field management, where interactions between injector and producer wells are inherently dynamic, it is vital to uncover causal connections to optimize recovery and minimize waste. Since controlled experiments are impractical in this setting, we must rely solely on observed data. In this paper, we develop an innovative causality-inspired framework that leverages domain expertise for causal feature learning for robust connectivity estimation. We address the challenge posed by confounding factors, latency in system responses, and the complexity of inter-well interactions that complicate causal analysis. First, we frame the problem through a causal lens and propose a novel framework that generates pairwise features driven by causal theory. This method captures meaningful representations of relationships within the oil field system. By constructing independent pairwise feature representations, our method implicitly accounts for confounder signal and enhances the reliability of connectivity estimation. Furthermore, our approach requires only limited context data to train machine learning models that estimate the connectivity probability between injectors and producers. We first validate our methodology through experiments on synthetic and semi-synthetic datasets, ensuring its robustness across varied scenarios. We then apply it to the complex Brazilian Pre-Salt oil fields using public synthetic and real-world data. Our results show that the proposed method effectively identifies injector-producer connectivity while maintaining rapid training times. This enables scalability and provides an interpretable approach for complex dynamic systems through causal theory. While previous projects have employed causal methods in the oil field context, to the best of our knowledge, this is the first time to systematically formulate the problem using causal reasoning that explicitly accounts for relevant confounders and develops an approach that effectively addresses these challenges and facilitates the discovery of interwell connections within an oil field.<\/jats:p>","DOI":"10.3389\/frai.2025.1686750","type":"journal-article","created":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T06:24:44Z","timestamp":1768458284000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Causality-driven feature representation for connectivity prediction"],"prefix":"10.3389","volume":"8","author":[{"given":"Bruno","family":"Souza","sequence":"first","affiliation":[{"name":"Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas","place":["Campinas, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Manuel","family":"Castro","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas","place":["Campinas, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ahmed","family":"Esmin","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas","place":["Campinas, Brazil"]},{"name":"Department of Computer Science, Federal University of Lavras (UFLA)","place":["Lavras, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Leonardo","family":"Machado","sequence":"additional","affiliation":[{"name":"Shell","place":["Rio de Janeiro, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alexandre","family":"Ferreira","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas","place":["Campinas, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anderson","family":"Rocha","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas","place":["Campinas, Brazil"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,1,15]]},"reference":[{"key":"B1","volume-title":"Information Theory","author":"Ash","year":"2012"},{"key":"B2","doi-asserted-by":"publisher","first-page":"767","DOI":"10.1613\/jair.1.13428","article-title":"Survey and evaluation of causal discovery methods for time series","volume":"73","author":"Assaad","year":"2022","journal-title":"J. Artif. Intell. Res"},{"key":"B3","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1145\/3501714.3501743","article-title":"\u201cOn pearl's hierarchy and the foundations of causal inference,\u201d","author":"Bareinboim","year":"2022","journal-title":"Probabilistic and Causal Inference: The Works of Judea Pearl"},{"key":"B4","article-title":"An introduction to causal reinforcement learning","author":"Bareinboim","year":"2021","journal-title":"arXiv preprint arXiv:2101.06498"},{"key":"B5","first-page":"884","article-title":"\u201cTime series deconfounder: estimating treatment effects over time in the presence of hidden confounders,\u201d","volume-title":"International Conference on Machine Learning","author":"Bica","year":"2020"},{"key":"B6","first-page":"3952","article-title":"\u201cInvariant causal imitation learning for generalizable policies,\u201d","author":"Bica","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B7","first-page":"37184","article-title":"\u201cTransfer learning on heterogeneous feature spaces for treatment effects estimation,\u201d","author":"Bica","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B8","doi-asserted-by":"publisher","first-page":"11402","DOI":"10.1038\/s41598-023-37929-w","article-title":"Time series causal relationships discovery through feature importance and ensemble models","volume":"13","author":"Castro","year":"2023","journal-title":"Sci. Rep"},{"key":"B9","article-title":"Unsupervised discovery of el nino using causal feature learning on microlevel climate data","author":"Chalupka","year":"2016","journal-title":"arXiv preprint arXiv:1605.09370"},{"key":"B10","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1007\/s41237-016-0008-2","article-title":"Causal feature learning: an overview","volume":"44","author":"Chalupka","year":"2017","journal-title":"Behaviormetrika"},{"key":"B11","first-page":"507","article-title":"Optimal structure identification with greedy search","volume":"3","author":"Chickering","year":"2002","journal-title":"J. Mach. Learn. Res"},{"key":"B12","article-title":"\u201cUnisim-ii-d: benchmark case proposal based on a carbonate reservoir,\u201d","volume-title":"SPE Latin America and Caribbean Petroleum Engineering Conference","author":"Correia","year":"2015"},{"key":"B13","first-page":"131","article-title":"On causal discovery from time series data using FCI","volume":"16","author":"Entner","year":"2010","journal-title":"Probab. Graph. Models"},{"key":"B14","doi-asserted-by":"publisher","first-page":"958","DOI":"10.1038\/s41591-024-02902-1","article-title":"Causal machine learning for predicting treatment outcomes","volume":"30","author":"Feuerriegel","year":"2024","journal-title":"Nat. Med"},{"key":"B15","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1515\/jci-2021-0048","article-title":"Causal inference in ai education: a primer","volume":"10","author":"Forney","year":"2022","journal-title":"J. Causal Infer"},{"key":"B16","doi-asserted-by":"publisher","first-page":"450","DOI":"10.1080\/01621459.2011.643719","article-title":"Evaluating the effect of training on wages in the presence of noncompliance, nonemployment, and missing outcome data","volume":"107","author":"Frumento","year":"2012","journal-title":"J. Am. Stat. Assoc"},{"key":"B17","volume-title":"Discovering Causal Structure: Artificial Intelligence, Philosophy of Science, and Statistical Modeling","author":"Glymour","year":"2014"},{"key":"B18","doi-asserted-by":"publisher","first-page":"424","DOI":"10.2307\/1912791","article-title":"Investigating causal relations by econometric models and cross-spectral methods","volume":"37","author":"Granger","year":"1969","journal-title":"Econometrica"},{"key":"B19","volume-title":"Propensity Score Analysis: Statistical Methods and Applications","author":"Guo","year":"2014"},{"key":"B20","doi-asserted-by":"publisher","first-page":"5507","DOI":"10.1175\/JCLI-D-17-0304.1","article-title":"Probabilities of causation of climate changes","volume":"31","author":"Hannart","year":"2018","journal-title":"J. Clim"},{"key":"B21","article-title":"A survey on causal discovery methods for iid and time series data","author":"Hasan","year":"2023","journal-title":"arXiv preprint arXiv:2303.15027"},{"key":"B22","first-page":"97","article-title":"\u201cSparse causal discovery in multivariate time series,\u201d","volume-title":"Causality: Objectives and Assessment","author":"Haufe","year":"2010"},{"key":"B23","author":"Hern\u00e1n","year":"2010","journal-title":"Causal Inference"},{"key":"B24","first-page":"3679","article-title":"\u201cCausal identification under markov equivalence: calculus, algorithm, and completeness,\u201d","author":"Jaber","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B25","first-page":"2981","article-title":"\u201cCausal identification under Markov equivalence: completeness results,\u201d","volume-title":"International Conference on Machine Learning","author":"Jaber","year":""},{"key":"B26","first-page":"32","article-title":"\u201cIdentification of conditional causal effects under Markov equivalence,\u201d","author":"Jaber","year":"","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B27","doi-asserted-by":"publisher","first-page":"26813","DOI":"10.1609\/aaai.v39i25.34885","article-title":"\u201cTesting causal models with hidden variables in polynomial delay via conditional independencies,\u201d","author":"Jeong","year":"2025","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"B28","first-page":"1404","article-title":"\u201cWell interference diagnosis through integrated analysis of tracer and pressure interference tests,\u201d","volume-title":"Unconventional Resources Technology Conference, Houston, Texas, 23\u201325 July 2018","author":"Kumar","year":"2018"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1623","DOI":"10.2118\/201233-PA","article-title":"Integrated analysis of tracer and pressure-interference tests to identify well interference","volume":"25","author":"Kumar","year":"2020","journal-title":"SPE J"},{"key":"B30","first-page":"23806","article-title":"\u201cMiracle: causally-aware imputation via learning missing data mechanisms,\u201d","author":"Kyono","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B31","first-page":"1501","article-title":"\u201cCastle: regularization via auxiliary causal graph discovery,\u201d","author":"Kyono","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B32","article-title":"\u201cEfficient neural causal discovery without acyclicity constraints,\u201d","author":"Lippe","year":"2022","journal-title":"International Conference on Learning Representations"},{"key":"B33","first-page":"43423","article-title":"\u201cCausal discovery from subsampled time series with proxy variables,\u201d","author":"Liu","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B34","doi-asserted-by":"publisher","first-page":"j4587","DOI":"10.1136\/bmj.j4587","article-title":"Handling time varying confounding in observational research","volume":"359","author":"Mansournia","year":"2017","journal-title":"BMJ"},{"key":"B35","doi-asserted-by":"publisher","first-page":"19","DOI":"10.3390\/make1010019","article-title":"Causal discovery with attention-based convolutional neural networks","volume":"1","author":"Nauta","year":"2019","journal-title":"Mach. Learn. Knowl. Extr"},{"key":"B36","doi-asserted-by":"publisher","first-page":"e1356","DOI":"10.1002\/widm.1356","article-title":"Bias in data-driven artificial intelligence systems\u2014an introductory survey","volume":"10","author":"Ntoutsi","year":"2020","journal-title":"Data Min. Knowl. Disc"},{"key":"B37","first-page":"1595","article-title":"\u201cDynotears: structure learning from time-series data,\u201d","volume-title":"International Conference on Artificial Intelligence and Statistics","author":"Pamfil","year":"2020"},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511803161","author":"Pearl","year":"2009","journal-title":"Causality"},{"key":"B39","first-page":"26","article-title":"\u201cCausal inference on time series using restricted structural equation models,\u201d","author":"Peters","year":"2013","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B40","doi-asserted-by":"publisher","first-page":"2553","DOI":"10.1038\/s41467-019-10105-3","article-title":"Inferring causation from time series in earth system sciences","volume":"10","author":"Runge","year":"","journal-title":"Nat. Commun"},{"key":"B41","doi-asserted-by":"publisher","first-page":"eaau4996","DOI":"10.1126\/sciadv.aau4996","article-title":"Detecting and quantifying causal associations in large nonlinear time series datasets","volume":"5","author":"Runge","year":"","journal-title":"Sci. Adv"},{"key":"B42","doi-asserted-by":"publisher","first-page":"853","DOI":"10.2118\/114983-PA","article-title":"Field applications of capacitance-resistance models in waterfloods","volume":"12","author":"Sayarpour","year":"2009","journal-title":"SPE Reser. Eval. Eng"},{"key":"B43","doi-asserted-by":"publisher","first-page":"612","DOI":"10.1109\/JPROC.2021.3058954","article-title":"Toward causal representation learning","volume":"109","author":"Sch\u00f6lkopf","year":"2021","journal-title":"Proc. IEEE"},{"key":"B44","doi-asserted-by":"publisher","first-page":"1759","DOI":"10.1109\/TSP.2015.2500893","article-title":"Granger causality in multivariate time series using a time-ordered restricted vector autoregressive model","volume":"64","author":"Siggiridou","year":"2015","journal-title":"IEEE Trans. Signal Proc"},{"key":"B45","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1177\/089443939100900106","article-title":"An algorithm for fast recovery of sparse causal graphs","volume":"9","author":"Spirtes","year":"1991","journal-title":"Soc. Sci. Comput. Rev"},{"key":"B46","author":"Spirtes","year":"2000","journal-title":"Constructing Bayesian network models of gene expression networks from microarray data"},{"key":"B47","first-page":"1","article-title":"\u201cCausal discovery and inference: concepts and recent methodological advances,\u201d","volume-title":"Applied Informatics","author":"Spirtes","year":"2016"},{"key":"B48","article-title":"NTS-notears: learning nonparametric DBNS with prior knowledge","author":"Sun","year":"2021","journal-title":"arXiv preprint arXiv:2109.04286"},{"key":"B49","first-page":"22221","article-title":"\u201cDECAF: generating fair synthetic data using causally-aware generative networks,\u201d","author":"Van Breugel","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B50","article-title":"Causal inference using llm-guided discovery","author":"Vashishtha","year":"2023","journal-title":"arXiv preprint arXiv:2310.15117"},{"key":"B51","doi-asserted-by":"publisher","first-page":"816","DOI":"10.3390\/en12050816","article-title":"Improved CRM model for inter-well connectivity estimation and production optimization: case study for karst reservoirs","volume":"12","author":"Wang","year":"2019","journal-title":"Energies"},{"key":"B52","first-page":"1437","article-title":"Causal reasoning with ancestral graphs","volume":"9","author":"Zhang","year":"2008","journal-title":"J. Mach. Learn. Res"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1686750\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T06:24:58Z","timestamp":1768458298000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1686750\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,15]]},"references-count":52,"alternative-id":["10.3389\/frai.2025.1686750"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1686750","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,15]]},"article-number":"1686750"}}