The seven tools of causal inference, with reflections on machine learning
Open accessAbstract
The kind of causal inference seen in natural human thought can be "algorithmitized" to help produce human-level machine intelligence.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Causal reasoning is foundational to human intelligence, and has been conjectured to be necessary for achieving human level AI.
“Causal reasoning is foundational to human intelligence, and has been conjectured to be necessary for achieving human level AI (Pearl, 2019).”
From Robust agents learn causal world models (2024) · cites this paper as Pearl, 2019
ML can play a role in causal inference, including in policy problems and in climate-relevant scenarios such as estimating the effects of temperature on human mortality and the effects of World Bank projects on vegetative cover.
“ML can play a role in causal inference [41,340,619], including in the context of policy problems [40,453] and in climate-relevant scenarios such as estimating the effects of temperature on human mortality [356] and the effects of World Bank projects on vegetative cover [883].”
From Tackling climate change with machine learning (2023) · cites this paper as [619]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241 · see citing works
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555 · see citing works
(1988) · Cited by 16,582 · see citing works
Self-citationAdvances in Neural Information Processing Systems 25
Zhang, Y., Sutton, C. A., Storkey, A., & Zoubin Ghahramani, A. (2012) · Cited by 12,653 · see citing works
Causality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653 · see citing works
Self-citationCAUSALITY: MODELS, REASONING AND INFERENCE
Hitchcock, C., & Pearl, J. (2001) · The Philosophical Review · doi:10.2307/3182612 · Cited by 4,824 · see citing works
Self-citationCausation, Prediction, and Search
Spirtes, P., Glymour, C., & Scheines, R. (2001) · doi:10.7551/mitpress/1754.001.0001 · Cited by 4,521 · see citing works
Causality: models, reasoning, and inference
(2000) · Choice Reviews Online · doi:10.5860/choice.38-1494a · Cited by 2,969 · see citing works
Human-level concept learning through probabilistic program induction
Lake, B. M., Salakhutdinov, R., & Tenenbaum, J. B. (2015) · Science · doi:10.1126/science.aab3050 · Cited by 2,817 · see citing works
Causal diagrams for empirical research
PEARL, J. (1995) · Biometrika · doi:10.1093/biomet/82.4.669 · Cited by 2,355 · cited 2× · see citing works
“…engine is available, called do - calculus , which predicts the effect of policy interventions whenever feasible, and exits with failure whenever predictions cannot be ascertained with the specified assumptions [Bareinboim and Pearl 2012; Pearl 1995; Shpitser and Pearl 2008; Tian and Pearl 2002].”
Self-citationIdentifiability and exchangeability for direct and indirect effects
Robins, J. M., & Greenland, S. (1992) · Epidemiology · doi:10.1097/00001648-199203000-00013 · Cited by 1,793 · see citing works
The Book of Why: The New Science of Cause and Effect
Goldberg, L. R. (2019) · Notices of the American Mathematical Society · doi:10.1090/noti1912 · Cited by 1,654 · see citing works
Proceedings of the 20th conference on Uncertainty in artificial intelligence
Meek, C., Chickering, M., & Halpern, J. Y. (2004) · Cited by 1,548 · see citing works
Explanation in Causal Inference: Methods for Mediation and Interaction
VanderWeele, T. J. (2015) · Cited by 1,540 · see citing works
van der Laan, M. J., & Rose, S. (2011) · Springer series in statistics · doi:10.1007/978-1-4419-9782-1 · Cited by 1,198 · see citing works
A Linear Non-Gaussian Acyclic Model for Causal Discovery
Shimizu, S., Hoyer, P. O., Hyvärinen, A., & Kerminen, A. (2006) · Cited by 1,152 · see citing works
Pearl, J. (2022) · ACM eBooks · doi:10.1145/3501714.3501736 · Cited by 1,039 · see citing works
Self-citationThe Book of Why: The New Science of Cause and Effect
Pearl, J., & Mackenzie, D. N. (2018) · Cited by 977 · see citing works
Self-citationDeep Learning: A Critical Appraisal
Marcus, G. (2018) · arXiv · doi:10.48550/arxiv.1801.00631 · Cited by 853 · see citing works
Graham, J. A. (2012) · Springer eBooks · doi:10.1007/978-1-4614-4018-5 · Cited by 799 · see citing works
Elements of Causal Inference: Foundations and Learning Algorithms
Peters, J., Janzing, D., & Schölkopf, B. (2017) · OAPEN (The OAPEN Foundation) · Cited by 794 · see citing works
Causes and Explanations: A Structural-Model Approach. Part I: Causes
Halpern, J. Y., & Pearl, J. (2005) · The British Journal for the Philosophy of Science · doi:10.1093/bjps/axi147 · Cited by 781 · see citing works
Self-citationCausal inference and the data-fusion problem
Bareinboim, E., & Pearl, J. (2016) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1510507113 · Cited by 621 · see citing works
Self-citationExternal Validity: From Do-Calculus to Transportability Across Populations
Pearl, J., & Bareinboim, E. (2014) · Statistical Science · doi:10.1214/14-sts486 · Cited by 455 · see citing works
Self-citationExternal validity: From do-calculus to transportability across populations
Pearl, J., & Bareinboim, E. (2013) · Cited by 398 · see citing works
Self-citationProceedings of the 16th Conference on Uncertainty in Artificial Intelligence
Boutilier, C. E., & Goldszmidt, M. (2000) · Cited by 350 · see citing works
[Bayesian Analysis in Expert Systems]: Comment: Graphical Models, Causality and Intervention
Pearl, J. (1993) · Statistical Science · doi:10.1214/ss/1177010894 · Cited by 258 · see citing works
Self-citationA general identification condition for causal effects
(2002) · eScholarship (California Digital Library) · doi:10.5555/777092.777180 · Cited by 203 · see citing works
Self-citationGraphical Models for Processing Missing Data
Mohan, K., & Pearl, J. (2021) · Journal of the American Statistical Association · doi:10.1080/01621459.2021.1874961 · Cited by 178 · see citing works
Self-citationComplete Identification Methods for the Causal Hierarchy
Shpitser, I., & Spirtes, P. (2007) · ePrints Soton (University of Southampton) · Cited by 136 · see citing works
Probabilistic Evaluation of Counterfactual Queries
Balke, A., & Pearl, J. (2022) · ACM eBooks · doi:10.1145/3501714.3501733 · Cited by 124 · see citing works
Self-citationGraphical Models for Inference with Missing Data
Mohan, K., Pearl, J., & Tian, J. W. (2013) · eScholarship (California Digital Library) · Cited by 107 · see citing works
Self-citationHuman-level intelligence or animal-like abilities?
Darwiche, A. Y. (2018) · Communications of the ACM · doi:10.1145/3271625 · Cited by 103 · see citing works
Causes of Effects and Effects of Causes
Pearl, J. (2014) · Sociological Methods & Research · doi:10.1177/0049124114562614 · Cited by 71 · see citing works
Self-citationCausal diagrams for empirical research
Pearl, J. (1995) · Biometrika · doi:10.1093/biomet/82.4.702 · Cited by 67 · see citing works
Self-citationTRYGVE HAAVELMO AND THE EMERGENCE OF CAUSAL CALCULUS
Pearl, J. (2014) · Econometric Theory · doi:10.1017/s0266466614000231 · Cited by 64 · see citing works
Self-citationComment: Graphical Models, Causality and Intervention
Pearl, J. (2016) · Cited by 62 · see citing works
Self-citationAI researchers allege that machine learning is alchemy
Hutson, M. (2018) · Science · doi:10.1126/science.aau0577 · Cited by 60 · see citing works
Causal Inference by Surrogate Experiments: z-IdentifiabilityPreprint
Bareinboim, E., & Pearl, J. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1210.4842 · Cited by 57 · see citing works
Self-citationGraphical Models, Causality, and InterventionPreprint
Pearl, J. (2011) · Cited by 40 · see citing works
Self-citationTransfer Learning in Multi-Armed Bandits: A Causal Approach
Zhang, J., & Bareinboim, E. (2017) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2017/186 · Cited by 38 · see citing works
Transfer Learning in Multi-Armed Bandit: A Causal Approach
Zhang, J., & Bareinboim, E. (2017) · Adaptive Agents and Multi-Agents Systems · doi:10.5555/3091125.3091436 · Cited by 30 · see citing works
Complete Identification Methods for the Causal Hierarchy
ShpitserIlya, & PearlJudea (2008) · Journal of Machine Learning Research · doi:10.5555/1390681.1442797 · Cited by 14 · see citing works
Identifiability and Exchangeability for Direct and
Robins, J. M., & Greenlandz, S. (1992) · Cited by 12 · see citing works
Causal Identification under Markov EquivalencePreprint
Jaber, A., Zhang, J., & Bareinboim, E. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.06209 · Cited by 4 · see citing works
Human-Level Intelligence or Animal-Like Abilities?
Darwiche, A. (2017) · arXiv · doi:10.48550/arxiv.1707.04327
Flagged first, then by citations
Research works printed but not linked: 3
- Morgan, S.L. and Winship, C. Counterfactuals and Causal Inference: Methods and Principles for Social Research (Analytical Methods for Social Research), Second Edition. Cambridge University Press, New York, 2015.
- Pearl, J. Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, San Mateo, CA, 1988.
- Morgan Kaufmann, San Francisco, CA, 2001, 411-420.
Cited by
Sorted by citations
Neurosymbolic AI: the 3rd wave
Garcez, A. D., & Lamb, L. C. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10448-w · Cited by 524 · see citing works
“In Pearl (2019), Pearl proposes a hierarchy consisting of three levels: association, intervention and counterfactual reasoning, and claims that ML is only capable of achieving association.”
BackgroundMachine learning and deep learning—A review for ecologists
Pichler, M., & Hartig, F. (2023) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14061 · Cited by 504 · see citing works
“Moreover, a pervasive concern is that ML models are trained for prediction, but the best predictive model does not necessarily correspond to the causal model (Breiman, 2001b; Pearl, 2019 , 2021; Box 4).”
Jung, J., Maeda, M. M., Chang, A., Bhandari, M., Ashapure, A., & Landivar-Bowles, J. (2020) · Current Opinion in Biotechnology · doi:10.1016/j.copbio.2020.09.003 · Cited by 456 · see citing works
Explainable AI Methods - A Brief Overview
Holzinger, A., Saranti, A., Molnar, C., Biecek, P., & Samek, W. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-04083-2_2 · Cited by 480 · see citing works
“Therefore, it will be important to measure the quality of explanations so that an xAI method achieves a certain level of causal understanding [65] for a user with effectiveness, efficiency and satisfaction in a given context of use”
MethodCRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories
Martínez‐Plumed, F., Contreras-Ochando, L., Ferri, C., Hernández‐Orallo, J., Kull, M., Lachiche, N., Ramírez-Quintana, M. J., & Flach, P. A. (2019) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2019.2962680 · Cited by 445 · see citing works
“Causal inference from data focuses on answering questions of the type “what if” and relies on methods that incorporate causal knowledge (such as the Structural Causal Models [46], the Potential Outcomes Framework [47] or the Linear non-Gaussian acyclic models [48]).”
MethodIntegrating explanation and prediction in computational social science
Hofman, J. M., Watts, D. J., Athey, S., Garip, F., Griffiths, T. L., Kleinberg, J., Margetts, H., Mullainathan, S., Salganik, M. J., Vazire, S., et al. (2021) · Nature · doi:10.1038/s41586-021-03659-0 · Cited by 371 · see citing works
Respecting causality for training physics-informed neural networks
Wang, S., Sankaran, S., & Perdikaris, P. G. (2024) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2024.116813 · Cited by 382 · see citing works
Holzinger, A., Malle, B., Saranti, A., & Pfeifer, B. (2021) · Information Fusion · doi:10.1016/j.inffus.2021.01.008 · Cited by 352 · see citing works
Chou, Y.-L., Moreira, C., Bruza, P., Ouyang, C., & Jorge, J. (2022) · Information Fusion · doi:10.1016/j.inffus.2021.11.003 · Cited by 302 · see citing works
Big Data in Earth system science and progress towards a digital twin
Li, X., Feng, M., Ran, Y., Su, Y., Liu, F., Huang, C., Shen, H., Xiao, Q., Su, J., Yuan, S., et al. (2023) · Nature Reviews Earth & Environment · doi:10.1038/s43017-023-00409-w · Cited by 269 · see citing works
“Confounding factors and selection bias can result in spurious correlations, which are ubiquitous in the universal relation network of machine learning, affecting their robustness, stability and interpretability 116 .”
BackgroundCausal interpretability for machine learning-problems, methods and evaluation
Moraffah, R., Karami, M., Guo, R., Raglin, A., & Liu, H. (2020) · ACM SIGKDD Explorations Newsletter · doi:10.1145/3400051.3400058 · Cited by 238 · see citing works
A literature review of machine learning algorithms for crash injury severity prediction
Santos, K., Dias, J. P., & Amado, C. (2021) · Journal of Safety Research · doi:10.1016/j.jsr.2021.12.007 · Cited by 236 · see citing works
Socially Responsible AI Algorithms: Issues, Purposes, and Challenges
Cheng, L., Varshney, K. R., & Liu, H. (2021) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.12814 · Cited by 261 · see citing works
Toward verified artificial intelligence
Seshia, S. A., Sadigh, D., & Sastry, S. S. (2022) · Communications of the ACM · doi:10.1145/3503914 · Cited by 215 · see citing works
Tackling climate change with machine learning
Rolnick, D., Donti, P. L., Kaack, L. H., Kochanski, K., Lacoste, A., Sankaran, K., Ross, A. S., Milojevic-Dupont, N., Jaques, N., Waldman-Brown, A., et al. (2023) · ACM Computing Surveys · doi:10.1145/3485128 · Cited by 336 · see citing works
Bickman, L. (2020) · Administration and Policy in Mental Health and Mental Health Services Research · doi:10.1007/s10488-020-01065-8 · Cited by 184 · see citing works
“I have found many individual papers that describe how to solve the causality problem with AI (e.g., Kuang et al. 2020; Pearl 2019).”
BackgroundArtificial Intelligence and Democracy: A Conceptual Framework
Jungherr, A. M. (2023) · Social Media + Society · doi:10.1177/20563051231186353 · Cited by 217 · see citing works
“This points to two problems: unobserved temporal shifts between variables (Lazer et al., 2014) and the dangers of relying on purely correlative evidence without the support of causal models (Pearl, 2019; Schölkopf et al., 2021).”
BackgroundHigh-throughput experimentation meets artificial intelligence: a new pathway to catalyst discovery
McCullough, K., Williams, T., Mingle, K., Jamshidi, P., & Lauterbach, J. A. (2020) · Physical Chemistry Chemical Physics · doi:10.1039/d0cp00972e · Cited by 180 · see citing works
Zhou, B., Li, X., Liu, T., Xu, K., Liu, W., & Bao, J. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2023.102333 · Cited by 181 · see citing works
A manifesto on explainability for artificial intelligence in medicine
Combi, C., Amico, B., Bellazzi, R., Holzinger, A., Moore, J. H., Žitnik, M., & Holmes, J. H. (2022) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2022.102423 · Cited by 177 · see citing works
Wang, Y. (2020) · Studies In Educational Evaluation · doi:10.1016/j.stueduc.2020.100872 · Cited by 196 · see citing works
“As a sub-set of AI, machine learning algorithms often use statistical techniques to adjust the algorithms to the situation and “learn” with data without being explicitly programmed to do so (Littman, 2015; Pearl, 2019).”
Reinforcement Learning in Economics and Finance
Arthur Charpentier, A., Élie, R., & Remlinger, C. (2021) · Computational Economics · doi:10.1007/s10614-021-10119-4 · Cited by 164 · see citing works
“Pearl (2019) asked the simple economic question ‘what will happen if we double the price’ (of an item we try to sell)?”
BackgroundExploring artificial intelligence for applications of drones in forest ecology and management
Buchelt, A., Adrowitzer, A., Kieseberg, P., Gollob, C., Nothdurft, A., Eresheim, S., Tschiatschek, S., Stampfer, K., & Holzinger, A. (2023) · Forest Ecology and Management · doi:10.1016/j.foreco.2023.121530 · Cited by 160 · see citing works
Marcus, J. L., Irie, W. C., Balzer, L. B., & Krakower, D. (2020) · Current HIV/AIDS Reports · doi:10.1007/s11904-020-00490-6 · Cited by 150 · see citing works
Taylor, J. E. T., & Taylor, G. W. (2020) · Psychonomic Bulletin & Review · doi:10.3758/s13423-020-01825-5 · Cited by 135 · see citing works
“To this end, we argue that explanations must be causal, rather than correlative (Pearl, 2019).”
BackgroundHariyanto, H., Kristianingsih, F. X. D., & Maharani, R. (2025) · Discover Education · doi:10.1007/s44217-025-00908-6 · Cited by 125 · see citing works
“Both supervised and unsupervised learning models underpin key functions such as learner profiling, content recommendation, and performance forecasting [28, 29].”
Liu, Y., Kelley, K. P., Vasudevan, R. K., Funakubo, H., Ziatdinov, M. A., & Kalinin, S. V. (2022) · Nature Machine Intelligence · doi:10.1038/s42256-022-00460-0 · Cited by 121 · see citing works
Machine learning for high-throughput experimental exploration of metal halide perovskites
Ahmadi, M., Ziatdinov, M., Zhou, Y.-Y., Lass, E., & Kalinin, S. V. (2021) · Joule · doi:10.1016/j.joule.2021.10.001 · Cited by 103 · see citing works
Automating the design-build-test-learn cycle towards next-generation bacterial cell factories
Gurdo, N., Volke, D. C., McCloskey, D., & Nikel, P. I. (2023) · New Biotechnology · doi:10.1016/j.nbt.2023.01.002 · Cited by 117 · see citing works
Shadi, M. R., Mirshekali, H., & Shaker, H. R. (2025) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2025.115668 · Cited by 127 · see citing works
Applications of artificial intelligence in drug development using real-world data
Chen, Z., Liu, X., Hogan, W. R., Shenkman, E. A., & Bian, J. (2020) · Drug Discovery Today · doi:10.1016/j.drudis.2020.12.013 · Cited by 108 · see citing works
“Causal modeling can also lead to improvements in the interpretability and adaptability of AI models in these drug development studies [45].”
BackgroundMachine learning for financial forecasting, planning and analysis: recent developments and pitfalls
Wasserbacher, H., & Spindler, M. (2021) · Digital Finance · doi:10.1007/s42521-021-00046-2 · Cited by 102 · see citing works
“The current lack of understanding of cause-effect connections is even cited as a fundamental obstacle for machine learning by some authors [40].”
BackgroundGradient-Based Neural DAG LearningPreprint
Lachapelle, S., Brouillard, P., Deleu, T., & Lacoste-Julien, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.02226 · Cited by 99 · see citing works
“Causal graphical models (CGM) [12] are BNs which support interventional queries like: What will happen if someone external to the system intervene on variable X? Recent work suggests that causality could partially solve challenges faced by current machine learning systems such as robustness to out-of-distribution samp…”
BackgroundToward Human–AI Interfaces to Support Explainability and Causability in Medical AI
Holzinger, A., & Müller, H. (2021) · Computer · doi:10.1109/mc.2021.3092610 · Cited by 95 · see citing works
Deep Structural Causal Models for Tractable Counterfactual InferencePreprint
Pawlowski, N., Castro, D. C., & Glocker, B. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.06485 · Cited by 96 · see citing works
“Second, we compare our work to recent progress in deep causal learning in light of Pearl’s ladder of causation [19].”
BackgroundResultAssociation is not prediction: A landscape of confused reporting in diabetes – A systematic review
Varga, T. V., Niss, K., Estampador, A. C., Collin, C. B., & Moseley, P. L. (2020) · Diabetes Research and Clinical Practice · doi:10.1016/j.diabres.2020.108497 · Cited by 94 · see citing works
Sahoh, B., & Choksuriwong, A. (2023) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-023-04594-w · Cited by 98 · see citing works
Preventing dataset shift from breaking machine-learning biomarkers
Dockès, J., Varoquaux, G. P., & Poline, J. (2021) · GigaScience · doi:10.1093/gigascience/giab055 · Cited by 86 · see citing works
“Unfortunately, the biases in participant recruitment may be similar in independently collected datasets.”
Quantifying Causal Pathways of Teleconnections
Kretschmer, M., Adams, S. V., Arribas, A., Prudden, R., Robinson, N., Saggioro, E., & Shepherd, T. G. (2021) · Bulletin of the American Meteorological Society · doi:10.1175/bams-d-20-0117.1 · Cited by 88 · see citing works
Cabitza, F., Campagner, A., Malgieri, G., Natali, C., Schneeberger, D., Stoeger, K., & Holzinger, A. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.118888 · Cited by 89 · see citing works
Explaining Black-Box Algorithms Using Probabilistic Contrastive Counterfactuals
Galhotra, S., Pradhan, R., & Salimi, B. (2021) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3448016.3458455 · Cited by 87 · see citing works
Causality-Based Neural Network Repair
Sun, B.-J., Sun, J., Pham, L. H., & Shi, T. (2022) · International Conference on Software Engineering · doi:10.1145/3510003.3510080 · Cited by 125 · see citing works
“SCM serves as a framework for representing what we know about the world, articulating what we want to know and connecting the two together in a solid semantics [49].”
BackgroundAn optimization approach for multi-echelon supply chain viability with disruption risk minimization
Liu, M., Liu, Z., Chu, F., Dolgui, A. B., Chu, C., & Zheng, F. (2022) · Omega · doi:10.1016/j.omega.2022.102683 · Cited by 79 · see citing works
“The do-calculus approach has been widely applied in transportation (Chatterjee et al., 2019), machine learning (Barber, 2012; Brownlee et al., 2022), causal inference (Zeˇcevi´c et al., 2021), and econometrics (Pearl, 2015; Pearl, 2019; Heckman and Pinto, 2022).”
Petersen, E., Potdevin, Y., Mohammadi, E., Zidowitz, S., Breyer, S., Nowotka, D., Henn, S., Pechmann, L., Leucker, M., Rostalski, P., et al. (2022) · IEEE Access · doi:10.1109/access.2022.3178382 · Cited by 83 · see citing works
“such as Bayesian networks— including, in particular, causal models [186], [323]—are interpretable as well, at least up to a certain degree of model complexity.”
BackgroundLINDA-BN: An interpretable probabilistic approach for demystifying black-box predictive models
Moreira, C., Chou, Y.-L., Velmurugan, M., Ouyang, C., Sindhgatta, R., & Bruza, P. (2021) · Decision Support Systems · doi:10.1016/j.dss.2021.113561 · Cited by 76 · see citing works
Causal GraphSAGE: A robust graph method for classification based on causal sampling
Zhang, T., Shan, H., & Little, M. A. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108696 · Cited by 75 · see citing works
Gonzales‐Inca, C., Calle, M., Croghan, D., Haghighi, A. T., Marttila, H., Silander, J., & Alho, P. (2022) · Water · doi:10.3390/w14142211 · Cited by 70 · see citing works
“Therefore, causal inferences might be questionable without robust assumptions and the veracity of the assumed data structure [237].”
BackgroundMethodGNN-SubNet: disease subnetwork detection with explainable graph neural networks
Pfeifer, B., Saranti, A., & Holzinger, A. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac478 · Cited by 69 · see citing works
Brain organoids and organoid intelligence from ethical, legal, and social points of view
Härtung, T., Pantoja, I. E. M., & Smirnova, L. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2023.1307613 · Cited by 66 · see citing works
“The development of causally-driven neural networks, which represent causal variables and relationships, is a promising direction (Pearl, 2019; Schölkopf, 2019).”
BackgroundShould Graph Convolution Trust Neighbors? A Simple Causal Inference Method
Feng, F., Huang, W., He, X., Xin, X., Wang, Q., & Chua, T. (2021) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3404835.3462971 · Cited by 65 · see citing works
“Apart from the new understanding of the conventional GCN inference, the causal theory provides analytical tools based on the causal graph, such as causal intervention [35], which enable the in-depth analysis of the factors resulting in the prediction and further reasoning based on the prediction [36].”
Trustworthy human-AI partnerships
Ramchurn, S. D., Stein, S., & Jennings, N. R. (2021) · iScience · doi:10.1016/j.isci.2021.102891 · Cited by 73 · see citing works
“In recent work, Pearl (2019) outlines a number of techniques for causal reasoning that may be able to learn causal relationships in data and traces.”
BackgroundDrug repurposing using real-world data
Tan, G. S. Q., Sloan, E. K., Lambert, P., Kirkpatrick, C. M. J., & Ilomäki, J. (2022) · Drug Discovery Today · doi:10.1016/j.drudis.2022.103422 · Cited by 63 · see citing works
Materials Data toward Machine Learning: Advances and Challenges
Zhu, L., Zhou, J., & Sun, Z. (2022) · The Journal of Physical Chemistry Letters · doi:10.1021/acs.jpclett.2c00576 · Cited by 63 · see citing works
Te’eni, D., Yahav, I., Zagalsky, A., Schwartz, D. G., Silverman, G., Cohen, D., Mann, Y., & Lewinsky, D. (2023) · Management Science · doi:10.1287/mnsc.2022.03518 · Cited by 61 · see citing works
Theory In, Theory Out: The Uses of Social Theory in Machine Learning for Social Science
Radford, J., & Joseph, K. (2020) · Frontiers in Big Data · doi:10.3389/fdata.2020.00018 · Cited by 58 · see citing works
“Specifying the theoretical model in advance is the only way to enumerate what features we should generate (Rohrer, 2018; Pearl, 2019).”
BackgroundRadical empiricism and machine learning research
Pearl, J. (2021) · Journal of Causal Inference · doi:10.1515/jci-2021-0006 · Cited by 60 · see citing works
“Such expectations are provided, for example, by causal models that predict both the outcomes of hypothetical manipulations as well the consequences of counterfactual undoing of past events (Pearl, 2019).”
BackgroundInfluentialSelf-citationTakhmasib, M., Lee, H. J., & Yi, H. (2023) · Automation in Construction · doi:10.1016/j.autcon.2023.105093 · Cited by 56 · see citing works
Improving Teamwork Competencies in Human-Machine Teams: Perspectives From Team Science
Stowers, K. L., Brady, L. L., MacLellan, C. J., Wohleber, R. W., & Eduardo Salas, A. (2021) · Frontiers in Psychology · doi:10.3389/fpsyg.2021.590290 · Cited by 57 · see citing works
“In contrast, as part of third wave AI, causal and counterfactual models (Pearl, 2019) aim to understand the causal relationships between variables.”
BackgroundThe Role of Causality in Explainable Artificial Intelligence
Carloni, G., Berti, A., & Colantonio, S. (2025) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.70015 · Cited by 73 · see citing works
“Pearl identifies some major obstacles still undermining the ability of AI systems in reasoning in a way akin to humans, to be overcome by equipping machines with causal modeling tools (Pearl, 2019).”
The need to move away from agential-AI: Empirical investigations, useful concepts and open issues
Cabitza, F., Campagner, A., & Simone, C. (2021) · International Journal of Human-Computer Studies · doi:10.1016/j.ijhcs.2021.102696 · Cited by 55 · see citing works
Hoogeveen, S., Haaf, J. M., Bulbulia, J. A., Ross, R., McKay, R. T., Altay, S., Bendixen, T., Berniūnas, R., Cheshin, A., Gentili, C., et al. (2022) · Nature Human Behaviour · doi:10.1038/s41562-021-01273-8 · Cited by 67 · see citing works
Exploring Evaluation Methods for Interpretable Machine Learning: A Survey
Alangari, N., Menaï, M. E. B., Mathkour, H. I., & Almosallam, I. A. (2023) · Information · doi:10.3390/info14080469 · Cited by 65 · see citing works
Accurate data-driven prediction does not mean high reproducibility
Li, J., Liu, L., Le, T. D., & Liu, J. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-019-0140-2 · Cited by 54 · see citing works
Black Boxes or Unflattering Mirrors? Comparative Bias in the Science of Machine Behaviour
Buckner, C. J. (2021) · The British Journal for the Philosophy of Science · doi:10.1086/714960 · Cited by 53 · see citing works
Clinical validation of saliency maps for understanding deep neural networks in ophthalmology
Ayhan, M. S., Kuemmerle, L. B., Kuehlewein, L., INHOFFEN, W., Aliyeva, G., Ziemssen, F., & Berens, P. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102364 · Cited by 52 · see citing works
From Deep Learning to Rational Machines
Buckner, C. J. (2023) · doi:10.1093/oso/9780197653302.001.0001 · Cited by 55 · see citing works
On the Binding Problem in Artificial Neural Networks
Greff, K., van Steenkiste, S., & Schmidhuber, J. (2020) · arXiv · doi:10.48550/arxiv.2012.05208 · Cited by 102 · see citing works
“…highly relevant for composition, due to their robustness under intervention and utility for reasoning about hypothetical or unobserved scenarios (Pearl, 2019; Schölkopf, 2019). of several layers that each produce an updated set of node representations by applying graph-convolutions to a local…”
BackgroundBiological Robots: Perspectives on an Emerging Interdisciplinary Field
Blackiston, D., Kriegman, S., Bongard, J., & Levin, M. (2022) · Soft Robotics · doi:10.1089/soro.2022.0142 · Cited by 48 · see citing works
“…optimum: state-of-the-art non-embodied AI is just as vulnerable to adversarial attacks or out-of-distribution environments as it always was, 59 and calls to incorporate causal reasoning into AI are growing 60 because non-embodied deep networks cannot cause effects and reason about the results.”
BackgroundGetting a CLUE: A Method for Explaining Uncertainty Estimates
Antorán, J., Bhatt, U., Adel, T., Weller, A., & Hernández-Lobato, J. M. (2021) · Apollo - University of Cambridge Repository · doi:10.17863/cam.69929 · Cited by 49 · see citing works
Lee, A., Inceoglu, I., Hauser, O., & Greene, M. (2022) · The Leadership Quarterly · doi:10.1016/j.leaqua.2020.101426 · Cited by 48 · see citing works
How much intelligence is there in artificial intelligence? A 2020 update
van der Maas, H. L. J., Snoek, L., & Stevenson, C. E. (2021) · Intelligence · doi:10.1016/j.intell.2021.101548 · Cited by 50 · see citing works
“Although Bayesian techniques are used more in psychology than in AI, many have argued that Bayesian modelling is crucial to overcoming the current limitations of AI (Lake, Ullman, Tenenbaum, & Gershman, 2017; Marcus, 2018; Pearl, 2019).”
Evaluating and Mitigating Bias in Image Classifiers: A Causal Perspective Using Counterfactuals
Dash, S., Balasubramanian, V. N., & Sharma, A. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00393 · Cited by 50 · see citing works
“, an image) based on a set of attributes A, Pearl’s ladder of causation [35] describes three types of distributions that can be used: association P (X|A = a), intervention P (X|do(A = a)) and counterfactual P (XA=a|A = a′, X = x′).”
BackgroundMethodSynthetic agency: sense of agency in artificial intelligence
Legaspi, R., He, Z., & Toyoizumi, T. (2019) · Current Opinion in Behavioral Sciences · doi:10.1016/j.cobeha.2019.04.004 · Cited by 47 · see citing works
The Way of the Future: Personalizing Treatment Plans Through Technology
Liefaard, M. C., Lips, E. H., Wesseling, J., Hylton, N. M., Lou, B., Mansi, T., & Pusztai, L. (2021) · American Society of Clinical Oncology Educational Book · doi:10.1200/edbk_320593 · Cited by 46 · see citing works
“58 Lastly, AI systems are required to provide insights and con fi dence estimates about their decision.”
Toward accountable human-centered AI: rationale and promising directions
Qadir, J., Islam, M., & Al-Fuqaha, A. I. (2022) · Journal of Information, Communication and Ethics in Society · doi:10.1108/jices-06-2021-0059 · Cited by 47 · see citing works
Big Data, Data Science, and Causal Inference: A Primer for Clinicians
Raita, Y., Camargo, C. A., Liang, L., & Hasegawa, K. (2021) · Frontiers in Medicine · doi:10.3389/fmed.2021.678047 · Cited by 46 · see citing works
“Despite these efforts, until recently, mainstream statistics has provided clinicians and researchers with few approaches to explicitly articulate, let alone to answer, causal questions (1, 8).”
BackgroundMethodCausal inference in drug discovery and development
Michoel, T., & Zhang, J. D. (2023) · Drug Discovery Today · doi:10.1016/j.drudis.2023.103737 · Cited by 47 · see citing works
“● Causal machine learning aiming at individualised causal inference using high-dimensional features to subdivide population, with emerging work in medicine and healthcare 43,151 .”
BackgroundGrand Challenges of Smart Technology for Older Adults
Zhou, J., Salvendy, G., Boot, W. R., Charness, N. H., Czaja, S. J., Gao, Q., Holzinger, A., Ntoa, S., Rau, P. P., Rogers, W. A., et al. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2457003 · Cited by 50 · see citing works
“One particularly relevant approach is the one of providing the possibility of asking questions of “why” (Pearl, 2019), and this can be further extended to a so-called counterfactual questioning (“what would happen if”) (Guidotti, 2024; Metsch et al., 2024; Prosperi et al., 2020; Roese, 1997;…”
Tolio, T. A. M., Monostori, L., Váncza, J., & Sauer, O. (2023) · CIRP Annals · doi:10.1016/j.cirp.2023.04.091 · Cited by 41 · see citing works
“Big data tends to be far more focused on correlation and misses causation [185].”
BackgroundThe Causal Fairness Field Guide: Perspectives From Social and Formal Sciences
Carey, A. N., & Wu, X. (2022) · AI and Ethics · doi:10.3389/fdata.2022.892837 · Cited by 90 · see citing works
“In Pearl (2019), Pearl presented the causal hierarchy through the Ladder of Causation, as shown in Figure 3 .”
BackgroundIskarous, K., & Pouplier, M. (2022) · Journal of Phonetics · doi:10.1016/j.wocn.2022.101195 · Cited by 37 · see citing works
Min, C., Wen, G., Gou, L., Li, X., & Yang, Z. (2023) · Energy · doi:10.1016/j.energy.2023.129211 · Cited by 40 · see citing works
Why We Should Teach Causal Inference: Examples in Linear Regression With Simulated Data
Lübke, K., Gehrke, M., Horst, J., & Szepannek, G. (2020) · Journal of Statistics Education · doi:10.1080/10691898.2020.1752859 · Cited by 37 · see citing works
Ketter, W., Padmanabhan, B., Pant, G., & Raghu, T. S. (2020) · Journal of the Association for Information Systems · doi:10.17705/1jais.00631 · Cited by 37 · see citing works
“Limitations of traditional predictive analytics are also leading researchers to seek casual frameworks for machine learning (Pearl, 2019).”
BackgroundGuide-target mismatch effects on dCas9–sgRNA binding activity in living bacterial cells
Feng, H., Guo, J., Wang, T., Zhang, C., & Xing, X. (2020) · Nucleic Acids Research · doi:10.1093/nar/gkaa1295 · Cited by 36 · see citing works
For a heterodox computational social science
Törnberg, P., & Uitermark, J. (2021) · Big Data & Society · doi:10.1177/20539517211047725 · Cited by 41 · see citing works
Mechanisms of Change in Exposure Therapy for Anxiety and Related Disorders: A Research Agenda
Benito, K., Pittig, A., Abramowitz, J. S., Arch, J. J., Chavira, D. A., de Kleine, R. A., Nadai, A. S. D., Hermans, D., Hofmann, S. G., Hoyer, J., et al. (2024) · Clinical Psychological Science · doi:10.1177/21677026241240727 · Cited by 37 · see citing works
“Modern approaches to specifying causal models and evaluating them statistically are heavily influenced by the counterfactual framework (Pearl, 2019b).”
Zhang, X., Zhang, T., Hou, L., Liu, X., Guo, Z., Tian, Y., & Liu, Y. (2025) · Systems · doi:10.3390/systems13070581 · Cited by 36 · see citing works
Degrees of algorithmic equivalence between the brain and its DNN models
Schyns, P. G., Snoek, L., & Daube, C. (2022) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2022.09.003 · Cited by 36 · see citing works
“, with which algorithmic method), although of course constraining any algorithm to process specific causal features is a strong constraint on what this algorithm must do [43].”
A systematic review of machine learning-based microscopic traffic flow models and simulations
Davies, R., He, H., Hui, F., Ali, Y., & Mohammed, Q. (2025) · Communications in Transportation Research · doi:10.1016/j.commtr.2025.100164 · Cited by 35 · see citing works
Empowering Language Understanding with Counterfactual Reasoning
Feng, F., Zhang, J., He, X., Zhang, H., & Chua, T. (2021) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2021.findings-acl.196 · Cited by 35 · see citing works
An Introduction to Causal InferencePreprint
Dablander, F. (2020) · doi:10.31234/osf.io/b3fkw · Cited by 35 · see citing works
“Following Pearl (2019b), I distinguish three “levels” of causal inference.”
BackgroundInfluentialQuantifying causality in data science with quasi-experiments
Liu, T., Ungar, L., & Kording, K. P. (2021) · Nature Computational Science · doi:10.1038/s43588-020-00005-8 · Cited by 34 · see citing works
Digital Twins in eHealth – : Prospects and Challenges Focussing on Information Management
Lutze, R. (2019) · IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) · doi:10.1109/ice.2019.8792622 · Cited by 33 · see citing works
“But, even for these techniques, AI pioneer Judea Pearl pointed out in a recent article [33] that most current AI techniques cover only the basic association level of causal reasoning.”
BackgroundMachine learning and structural econometrics: contrasts and synergies
Iskhakov, F., Rust, J. P., & Schjerning, B. (2020) · Econometrics Journal · doi:10.1093/ectj/utaa019 · Cited by 33 · see citing works
“Pearl (2019) also raised the question of robustness: if the environment that an actor–critic algorithm is ‘expecting’ changes sufficiently, will its performance break down and result in counterproductive forecasts and recommendations?”
BackgroundWhat would have happened? Counterfactuals, hypotheticals and causal judgements
Gerstenberg, T. (2022) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2021.0339 · Cited by 33 · see citing works
“To explain the pattern of causal judgements across the different video clips, a model requires the causal knowledge and computational capacities on level III of Pearl’s hierarchy.”
BackgroundInfluentialGenerative causal explanations of black-box classifiersPreprint
O’Shaughnessy, M. R., Canal, G., Connor, M., Davenport, M. A., & Rozell, C. J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.13913 · Cited by 33 · see citing works
“Because explanation methods seek to answer “why” and “how” questions that use the language of cause and effect [3, 4], causal reasoning has played an increasingly important role in designing explanation frameworks [5].”
BackgroundSocially Responsible AI Algorithms: Issues, Purposes, and ChallengesPreprint
Cheng, L., Varshney, K. R., & Liu, H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.02032 · Cited by 34 · see citing works
“Causal relationship, however, is expected to be invariant and robust across environments (Pearl, 2019, 2009).”
BackgroundYu, R., Zheng, Y., Abdel-Aty, M. A., & Gao, Z. (2019) · Accident Analysis & Prevention · doi:10.1016/j.aap.2019.01.022 · Cited by 32 · see citing works
Evans, J. (2020) · Journal of Social Computing · doi:10.23919/jsc.2020.0002 · Cited by 32 · see citing works
“This has led social scientists to recruit insights from machine learning into their quest for causal explanation [53] as many experts in machine learning have come to see causality as the next great frontier in their field [54] in a quest to yield simulations that effectively redesign society.”
Background
- Authors
- Pearl, J.
- Affiliations
- University of California, Los Angeles (US)
- Corresponding author
- Pearl, Judea
- Venue
- Communications of the ACM
- Published
- 2019-02-21
- Volume
- 62
- Issue
- 3
- Pages
- 54-60
- ISSN
- 0001-0782 (ISSN-L)
- 1557-7317
- OpenAlex
- W2917286209
- Cited by
- 659
- References
- 46
- Citations by year
- 2026: 78 · 2025: 81 · 2024: 91 · 2023: 98 · 2022: 102 · 2021: 118 · 2020: 67 · 2019: 24
- FWCI
- 35.88
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Communications of the ACM (publishedVersion)
- Funders
- National Science Foundation (1704932, 1302448, 1527490)
- Defense Advanced Research Projects Agency (#W911NF-16-057)
- University of California, Los Angeles
- Advanced Research Projects Agency
- Office of Naval Research (N00014-17-S-B001, N00014)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Physical Sciences › Computer Science › Artificial Intelligence › Logic, Reasoning, and Knowledge
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Keywords
- causal inference, causal discovery, causal discovery algorithms, machine learning, philosophy of causation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
