Next Article in Journal
Information-Driven Rule Reduction in Belief Rule Bases for Complex System Modeling
Previous Article in Journal
Image Encryption Algorithm Based on a Novel Hyperchaotic Map and 3D Histogram Model
Previous Article in Special Issue
An Algorithmic Treatment of Causal Unit Selection
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Entropy
Volume 28
Issue 5
10.3390/e28050577
entropy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

Causal Structure Learning Assumptions Shape Counterfactual Safety: Expert-Guided Constraints vs. Data-Driven DAGs with Probabilistic Logic Twin Networks †

by
Héctor Avilés
1,*,
Ingridh Gracia
1,
Rafael Kiesel
2,
Verónica Rodríguez
3,*,
Rubén Machucho
1,
Alberto Reyes
4,
Marco Negrete
5,
Gabriel Ramírez
6,
Nicolás Luévano
1,
Myriam Pequeño
1,
Jesús Medrano
1 and
Felix Weitkämper
7
1
Department of Information Technologies, Polytechnic University of Victoria, Ciudad Victoria, 87138, Mexico
2
Institute of Computational Logic, Vienna University of Technology, 1040 Vienna, Austria
3
Computer Science Institute, Technological University of the Mixteca, Huajuapan de León, 69004, Mexico
4
Control, Electronics and Communications Department, National Institute of Electricity and Clean Energies, Cuernavaca 62490, Mexico
5
Faculty of Engineering, National Autonomous University of Mexico, Mexico City 04510, Mexico
6
Center for Research and Advanced Studies, The National Polytechnic Institute, Tamaulipas Campus, Ciudad Victoria 87138, Mexico
7
Statistical and Symbolic Artificial Intelligence Group, German University of Digital Science, 14482 Potsdam, Germany
*
Authors to whom correspondence should be addressed.
†
This paper is an extended version of our paper published in Cadis Workshop 2025 (https://cadisworkshop.com.mx/)
Entropy 2026, 28(5), 577; https://doi.org/10.3390/e28050577
Submission received: 28 February 2026 / Revised: 1 May 2026 / Accepted: 18 May 2026 / Published: 21 May 2026
(This article belongs to the Special Issue Causal Graphical Models and Their Applications, 2nd Edition)
Versions Notes

Abstract

We investigate how causal DAG learning algorithms and structural assumptions influence counterfactual decision safety. Four structure learning regimes are compared: expert-guided edge-constrained HC+BIC, unconstrained HC+BIC, MMPC+HC+BIC, and the PC-Stable algorithm. Evaluation is conducted using a leave-one-state-out protocol over a fully enumerated state–action space in a controlled offline autonomous driving setting. The environment is characterized by seven Boolean state variables and six actions, allowing us to disentangle the effects of learning strategies on counterfactual decisions. All models are implemented as probabilistic logic twin networks (PLTNs), with additional sensitivity analysis across parameter configurations. The learning regimes produce markedly different counterfactual decisions. Edge-constrained HC+BIC recommends a diverse set of safe actions, while unconstrained HC+BIC yields fewer but consistently safe alternatives. MMPC+HC+BIC frequently fails to identify safe actions, often associated with weak connectivity of the outcome variable. PC-Stable produces varied recommendations but may include unsafe actions, which is linked to incorrect edge orientations between actions and outcomes. These findings show that structure learning choices and prior knowledge influence counterfactual decisions through the learned structure, affecting the identification of safe alternatives in safety-critical applications.
Keywords: causal discovery; counterfactual reasoning; probabilistic logic; autonomous systems causal discovery; counterfactual reasoning; probabilistic logic; autonomous systems

Share and Cite

MDPI and ACS Style

Avilés, H.; Gracia, I.; Kiesel, R.; Rodríguez, V.; Machucho, R.; Reyes, A.; Negrete, M.; Ramírez, G.; Luévano, N.; Pequeño, M.; et al. Causal Structure Learning Assumptions Shape Counterfactual Safety: Expert-Guided Constraints vs. Data-Driven DAGs with Probabilistic Logic Twin Networks. Entropy 2026, 28, 577. https://doi.org/10.3390/e28050577

AMA Style

Avilés H, Gracia I, Kiesel R, Rodríguez V, Machucho R, Reyes A, Negrete M, Ramírez G, Luévano N, Pequeño M, et al. Causal Structure Learning Assumptions Shape Counterfactual Safety: Expert-Guided Constraints vs. Data-Driven DAGs with Probabilistic Logic Twin Networks. Entropy. 2026; 28(5):577. https://doi.org/10.3390/e28050577

Chicago/Turabian Style

Avilés, Héctor, Ingridh Gracia, Rafael Kiesel, Verónica Rodríguez, Rubén Machucho, Alberto Reyes, Marco Negrete, Gabriel Ramírez, Nicolás Luévano, Myriam Pequeño, and et al. 2026. "Causal Structure Learning Assumptions Shape Counterfactual Safety: Expert-Guided Constraints vs. Data-Driven DAGs with Probabilistic Logic Twin Networks" Entropy 28, no. 5: 577. https://doi.org/10.3390/e28050577

APA Style

Avilés, H., Gracia, I., Kiesel, R., Rodríguez, V., Machucho, R., Reyes, A., Negrete, M., Ramírez, G., Luévano, N., Pequeño, M., Medrano, J., & Weitkämper, F. (2026). Causal Structure Learning Assumptions Shape Counterfactual Safety: Expert-Guided Constraints vs. Data-Driven DAGs with Probabilistic Logic Twin Networks. Entropy, 28(5), 577. https://doi.org/10.3390/e28050577

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop