Causal Structure Learning

Follow this topic Ask about this topic

Causal structure learning seeks graphs whose arrows represent direct cause-and-effect relationships, helping explain systems and predict what may change under interventions. Methods include conditional-independence tests such as PC and FCI, score-based and Bayesian graph search, functional models such as LiNGAM and additive noise models, and continuous optimization methods such as NOTEARS; interventions and temporal structure can further constrain the search. Work increasingly addresses time-varying systems, latent confounders, high-dimensional data, and ways to incorporate expert or semantic knowledge. A central challenge is that observational data often leave multiple graphs indistinguishable, while stronger identifying assumptions, hidden variables, noisy data, and scalability complicate reliable recovery and evaluation.

Latest papers

124 papers19 papers in the last four weeks, up 73% on the four weeks before. 0.1% of all new papers.

Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions?
Oct 25, 2026Mallesh MadapathiCausal World Models Action-Conditioned World Models
ThinkingDBx Pvt. Ltd., Hyderabad, India
Operationalising Weaker Faithfulness: Recovery, Ambiguity and Computational Limits
Oct 22, 2026Cillian Hourican, Rick QuaxConstraint-Based Causal Discovery Causal Discovery
Computational Science Lab, Informatics Institute, University of Amsterdam. · Computational Science Lab, Informatics Institute, University of Amsterdam
Learning Non-Gaussian SEMs with Conditional Risk, and Why the Direction of Recursion Matters
Oct 9, 2026Ujas Shah, Panagiota Birmpa, Rebecca MorrisonCausal Structure Learning Structural Causal Models
University of Colorado Boulder · Heriot-Watt University
One Thing Leads to Another: The Essentials of Time Series Causal Discovery
Oct 9, 2026Kurt Butler, Damian Machlanski, Michael P. J. Camilleri +5Time-Series Causal Discovery Causal Discovery
School of Engineering, The University of Edinburgh, and the Causality in Healthcare AI Hub (CHAI) · School of Informatics, The University of Edinburgh and CHAI · Department of Statistical Science, University College London, and CHAI
DAG Structure Learning with Node-Varying Heavy-Tailed Disturbances
Oct 9, 2026Robin Francis, Sundeep Prabhakar ChepuriCausal Structure Learning Causal Discovery
Indian Institute of Science, Bangalore, India
Score-Based Learning of Cluster DAGs from Interventions
Oct 8, 2026Gaetano Tedesco, Alex MarkhamScore-Based Causal Discovery Causal Structure Learning
Informatics Institute, University of Amsterdam · Department of Mathematical Sciences, University of Copenhagen
Executing Causal Structure Learning with Linear-Attention Transformers
Oct 7, 2026Amartya Roy, Sayar KarmakarNeural Algorithmic Reasoning Causal Structure Learning
School of Interdisciplinary Research Indian Institute of Technology Delhi · University of Florida
AdaPS-LiNGAM: Adaptive Predecessor Selection for Linear Non-Gaussian Acyclic Models under Small-Sample Settings
Oct 7, 2026Shun Yanashima, Kentaro Kanamori, Hirofumi SuzukiCausal Discovery Causal Structure Learning
Artiﬁcial Intelligence Laboratory, Fujitsu Limited, 4-1-1 Kamikodanaka, Nakahara-ku, Kawasaki, 211-8588, Kanagawa, Japan.
Directional Evidence Guided Search-Space Reduction for Exact DAG Learning
Oct 6, 2026Upala Junaida Islam, Abdelmonem Elrefaey, Rong PanScore-Based Causal Discovery Bayesian Network Structure Learning
School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85281, USA.
GO-Based Clustering for Learning Cluster-Level Causal Gene Regulatory Networks
Oct 5, 2026Azlaan Mustafa Samad, Wei Zhang, Adèle H RibeiroClustering Causal Discovery
L3S Research Center, Leibniz University Hannover, Lower Saxony Center for Artificial Intelligence and Causal Methods in Medicine (CAIMed), Hannover, Germany · Institute for Causal and Explainable Artificial Intelligence in Life Sciences, Center for Computational Life Sciences, RWTH Aachen University, Germany
Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discovery
Oct 3, 2026Guoliang Xu, James E CorterCausal Discovery Causal Structure Learning
Columbia University
Hold-Out Scoring for Efficient Gaussian DAG Learning
Oct 2, 2026Donguk Shin, Byeongguk Kang, Inseol Lee +1Graph Structure Learning Causal Structure Learning
Department of Statistics, Seoul National University. · Interdisciplinary Program in Artificial Intelligence, Seoul National University. · Institute for Data Innovation in Science, Seoul National University.
How Causality Bridges the Semantic Gap
Oct 1, 2026Shuhao Zhang, Xuran Zhou, Han Guo +2Semantic Mapping Causal Structure Learning
University of California, San Diego · University of Illinois Urbana-Champaign
Structure-agnostic Causal Representation Learning
Oct 1, 2026Arman Behnam, Binghui WangCausal Representation Learning OOD Generalization
Department of Computer Science Illinois Institute of Technology Chicago, IL, USA
Differentiable Structure Learning for Cyclic Linear Gaussian Models with Latent Confounders
Sep 29, 2026Sadegh Khorasani, Ali Najar, Saber Salehkaleybar +1Structural Causal Models Causal Discovery
School of Computer and Communication Sciences, EPFL, Lausanne, Switzerland · Leiden Institute of Advanced Computer Science (LIACS), Leiden University, Leiden, The Netherlands · College of Management of Technology, EPFL, Lausanne, Switzerland
Demistifying Data and Simulator Assumptions in Supervised Causal Discovery
Sep 29, 2026Pingchuan Ma, Rui Ding, Bojun Huang +1Causal Discovery Causal Structure Learning
Zhejiang University of Technology, China · Microsoft Research Asia · Sony Research +1
Do-JEPA: From Masking to Intervention in Latent World Models
Sep 29, 2026Hossein Resani, Javen Qinfeng ShiCausal Effect Estimation Causal World Models
Australian Institute for Machine Learning, Adelaide University · Responsible AI Research Centre
Forecast-Necessary Causal Discovery for Nonlinear Political Panel Data: Feedback, Functional Form, and the Dynamics of Democratization
Sep 28, 2026Michael Coppedge, Dmitry Zaytsev, Valentina KuskovaTime-Series Causal Discovery Causal Discovery
Department of Political Science, University of Notre Dame, Notre Dame, IN, USA · Kellogg Institute, University of Notre Dame, Notre Dame, IN, USA · Lucy Family Institute for Data & Society, University of Notre Dame, Notre Dame, IN, USA
The Statistical Cost of Causal Discovery with Feedback
Sep 28, 2026Sunmin Oh, Seungsu Han, Gunwoong ParkSample Complexity Causal Discovery
Department of Statistics; Institute for Data Innovation in Science, Seoul National University, Korea · Department of Operations Research and Financial Engineering, Princeton University, USA · Department of Statistics; Interdisciplinary Program in Artificial Intelligence; Institute for Data Innovation in Science, Seoul National University, Korea.
Feedback-Robust AI for Patient Knowledge Graphs
Sep 27, 2026Mohammed Sameer SyedBiomedical KGs KG Construction
Roshan AI
LUCID: Learning Under Confounding for Inference and Discovery in Time Series
Sep 25, 2026Mohammad FesangharyTime-Series Causal Discovery Causal Inference with Unobserved Confounding
Bloomberg L.P.
xWhyL: Causal Interactive Learning
Sep 22, 2026Nicholas Tagliapietra, Florian Peter Busch, Moritz Willig +4Causal Discovery Causal Structure Learning
Bosch Center for Artificial Intelligence, Germany · Computer Science Department, Technical University of Darmstadt, Germany · Hessian Center for AI (hessian.AI), Darmstadt, Germany
Hessian Rank Constraint for Learning Structure of Nonlinear Latent Variable Models
Sep 21, 2026Zijian Li, Ruichu Cai, Feng Xie +7Latent Variable Models Causal Discovery
Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE · Carnegie Mellon University, Pittsburgh, PA, USA · Guangdong University of Technology, Guangzhou, China +3
Decoupled Causal Discovery
Sep 20, 2026Zhengkang Guan, Fei Wu, Kun KuangCausal Discovery Causal Structure Learning
College of Computer Science and Technology Zhejiang University
MaSCoD: A Multi-Agent Framework for Structural-Context-Guided Candidate Causal Graph Generation
Sep 17, 2026Yudai Nakada, Yuichiro Nishiura, Jin Michael SplichalMulti-Agent LLM Systems Causal Discovery
SCSK Corporation Tokyo, Japan
Provable Guarantees and Efficient Learning of Structural Equation Models with Latent Confounders
Sep 16, 2026Weijian Yu, Jean HonorioStructural Causal Models Low-Rank Matrix Decomposition
CIS, The University of Melbourne
On the Identifiability of Mixed Ordinal and Exponential Family Causal DAGs under Linear Parametric Models
Sep 16, 2026Sambit Mishra, Urbashi MitraStructural Causal Models Causal Discovery
Ming Hsieh Department of Electrical and Computer Engineering University of Southern California Los Angeles, CA 90089, USA
Fundamental Dynamical Units for Physics-Informed Structural Inference from Perturbation Time-Series in Networked Systems
Sep 14, 2026Nima NouriTime-Series Causal Discovery Causal Discovery
Artificial Intelligence for Science Innovation, AstraZeneca, Waltham, MA, USA
Tensor Network Moral Graph Recovery of Discrete Probability Distributions
Sep 8, 2026Á. Troyano Olivas, Chi-Hang Fred Fung, Hans H. Brunner +2Tensor Networks Graph Structure Learning
Heisenberg Research Center (Munich), Huawei Technologies Duesseldorf GmbH, Germany · Center for Computational Simulation, Universidad Politécnica de Madrid, Madrid, Spain
Causal DAG Identification for Count Data via Poisson Thinning Structural Equation Models
Sep 5, 2026Penggang Gao, Ming Cai, Hisayuki HaraStructural Causal Models Causal Discovery
Graduate School of Informatics Kyoto University Kyoto, 606-8501, Japan · Institute for Liberal Arts and Sciences Kyoto University Kyoto, 606-8501, Japan

Oct 25, 2026J/K move · Enter open · S save

Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions?

Mallesh Madapathi

ThinkingDBx Pvt. Ltd., Hyderabad, India

Causal World Models Action-Conditioned World Models Model-Based Planning Causal Structure Learning

Open paper AskPDF arXiv

Agents that act on real systems should predict what their actions will cause before acting. We study when causal structure helps a learned forward model do this, using a planner that rejects candidate actions whose simulated multi-step outcome is unsafe, in synthetic structural causal models where the true consequences are known. We compare the Action-Conditioned Latent Structural Causal Model (AC-LSCM: per-factor mechanisms over a learned sparse graph, actions as do-interventions) with three unstructured forward models on seven random graphs with 10 or 20 variables, a task configured by a fixed rule, and two controls: a do-wrapper that gives every model the same action semantics, and counterfactual data given to all models or to none. When its factors are the observed variables, AC-LSCM recovers the true graph exactly on every 10-variable graph and has lower three-step prediction error than every baseline in all 396 paired comparisons. Its excess safety violations over an oracle are 0.25 of the no-model level on 10-variable graphs (best baseline 0.31) and 0.41 on 20-variable graphs (0.53), and its margin is largest for two-variable actions never seen in training (0.18 versus 0.28). With learned latent factors the same model is worse than the baselines; the choice of factors matters more than the explicit do-operator or counterfactual data, and every planner stays far from the oracle. This version corrects an earlier one, whose model learned an empty graph and whose headline 36x safety improvement came from a task that tested only extrapolation of action values; we withdraw that claim. Code and all per-seed results are released.