More Like this

1. Constraint-based Causal Discovery from a Collection of Conditioning Sets

   Lee, Kenneth; Ribeiro, Bruno; Kocaoglu, Murat (August 2025, Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence)

   In constraint-based causal discovery, the existing algorithms systematically use a series of conditional independence (CI) relations observed in the data to recover an equivalence class of causal graphs in the large sample limit. One limitation of these algorithms is that CI tests lose statistical power as conditioning set size increases with finite samples. Recent research proposes to limit the conditioning set size for robust causal discovery. However, the existing algorithms require exhaustive testing of all CI relations with conditioning set sizes up to a certain integer k. This becomes problematic in practice when variables with large support are present, as it makes CI tests less reliable due to near-deterministic relationships, thereby violating the faithfulness assumption. To address this issue, we propose a causal discovery algorithm that only uses CI tests where the conditioning sets are restricted to a given set of conditioning sets including the empty set C. We call such set of CI relations IC conditionally closed. We define the notion of C-Markov equivalence: two causal graphs are C-Markov equivalent if they entail the same set of CI constraints from IC. We propose a graphical representation of C-Markov equivalence and characterize such equivalence between two causal graphs. Our proposed algorithm called the C-PC algorithm is sound for learning the C-Markov equivalence class. We demonstrate the utility of the proposed algorithm via synthetic and real-world experiments in scenarios where variables with large support or high correlation are present in the data. Our source code is available online at github.com/kenneth-lee-ch/cpc. 

   more » « less
2. Identification of Linear Non-{G}aussian Latent Hierarchical Structure

   Xie, Feng; Huang, Biwei; Chen, Zhengming; He, Yangbo; Geng, Zhi; Zhang, Kun (July 2022, Proceedings of Machine Learning Research)

   Chaudhuri, Kamalika; Jegelka, Stefanie; Song, Le; Szepesvari, Csaba; Niu, Gang; Sabato, Sivan (Ed.)

   Traditional causal discovery methods mainly focus on estimating causal relations among measured variables, but in many real-world problems, such as questionnaire-based psychometric studies, measured variables are generated by latent variables that are causally related. Accordingly, this paper investigates the problem of discovering the hidden causal variables and estimating the causal structure, including both the causal relations among latent variables and those between latent and measured variables. We relax the frequently-used measurement assumption and allow the children of latent variables to be latent as well, and hence deal with a specific type of latent hierarchical causal structure. In particular, we define a minimal latent hierarchical structure and show that for linear non-Gaussian models with the minimal latent hierarchical structure, the whole structure is identifiable from only the measured variables. Moreover, we develop a principled method to identify the structure by testing for Generalized Independent Noise (GIN) conditions in specific ways. Experimental results on both synthetic and real-world data show the effectiveness of the proposed approach. 

   more » « less
3. Differentiable Cyclic Causal Discovery Under Unmeasured Confounders

   Sethuraman, Muralikrishnna G; Fekri, Faramarz (December 2025, 39th Conference on Neural Information Processing Systems (NeurIPS 2025))

   Understanding causal relationships between variables is fundamental across scientific disciplines. Most causal discovery algorithms rely on two key assump- tions: (i) all variables are observed, and (ii) the underlying causal graph is acyclic. While these assumptions simplify theoretical analysis, they are often violated in real-world systems, such as biological networks. Existing methods that account for confounders either assume linearity or struggle with scalability. To address these limitations, we propose DCCD-CONF, a novel framework for differentiable learning of nonlinear cyclic causal graphs in the presence of unmeasured confounders using interventional data. Our approach alternates between optimizing the graph structure and estimating the confounder distribution by maximizing the log-likelihood of the data. Through experiments on synthetic data and real-world gene perturbation datasets, we show that DCCD-CONF outperforms state-of-the-art methods in both causal graph recovery and confounder identification. Additionally, we provide consistency guarantees for our framework, reinforcing its theoretical soundness. 

   more » « less
4. KCRL: A Prior Knowledge Based Causal Discovery Framework With Reinforcement Learning

   Hasan, Uzma; Gani, Md Osman (April 2022, Proceedings of Machine Learning Research)

   Causal discovery is an important problem in many sciences that enables us to estimate causal relationships from observational data. Particularly, in the healthcare domain, it can guide practitioners in making informed clinical decisions. Several causal discovery approaches have been developed over the last few decades. The success of these approaches mostly relies on a large number of data samples. In practice, however, an infinite amount of data is never available. Fortunately, often we have some prior knowledge available from the problem domain. Particularly, in healthcare settings, we often have some prior knowledge such as expert opinions, prior RCTs, literature evidence, and systematic reviews about the clinical problem. This prior information can be utilized in a systematic way to address the data scarcity problem. However, most of the existing causal discovery approaches lack a systematic way to incorporate prior knowledge during the search process. Recent advances in reinforcement learning techniques can be explored to use prior knowledge as constraints by penalizing the agent for their violations. Therefore, in this work, we propose a framework KCRL that utilizes the existing knowledge as a constraint to penalize the search process during causal discovery. This utilization of existing information during causal discovery reduces the graph search space and enables a faster convergence to the optimal causal mechanism. We evaluated our framework on benchmark synthetic and real datasets as well as on a real-life healthcare application. We also compared its performance with several baseline causal discovery methods. The experimental findings show that penalizing the search process for constraint violation yields better performance compared to existing approaches that do not include prior knowledge. 

   more » « less
5. CausalSim: A Causal Framework for Unbiased Trace-Driven Simulation

   Abdullah Alomar, Pouya Hamadanian (April 2023, Proceedings of the 20th USENIX Symposium on Networked Systems Design and Implementation)

   We present CausalSim, a causal framework for unbiased trace-driven simulation. Current trace-driven simulators assume that the interventions being simulated (e.g., a new algorithm) would not affect the validity of the traces. However, real-world traces are often biased by the choices algorithms make during trace collection, and hence replaying traces under an intervention may lead to incorrect results. CausalSim addresses this challenge by learning a causal model of the system dynamics and latent factors capturing the underlying system conditions during trace collection. It learns these models using an initial randomized control trial (RCT) under a fixed set of algorithms, and then applies them to remove biases from trace data when simulating new algorithms. Key to CausalSim is mapping unbiased trace-driven simulation to a tensor completion problem with extremely sparse observations. By exploiting a basic distributional invariance property present in RCT data, CausalSim enables a novel tensor completion method despite the sparsity of observations. Our extensive evaluation of CausalSim on both real and synthetic datasets, including more than ten months of real data from the Puffer video streaming system shows it improves simulation accuracy, reducing errors by 53% and 61% on average compared to expert-designed and supervised learning baselines. Moreover, CausalSim provides markedly different insights about ABR algorithms compared to the biased baseline simulator, which we validate with a real deployment 

   more » « less