Your Assumed DAG is Wrong and Here's How To Deal With It
Not on the map. Why?
Abstract (arXiv)
Assuming a directed acyclic graph (DAG) that represents prior knowledge of causal relationships between variables is a common starting point for cause-effect estimation. Existing literature typically invokes hypothetical domain expert knowledge or causal discovery algorithms to justify this assumption. In practice, neither may propose a single DAG with high confidence. Domain experts are hesitant to rule out dependencies with certainty or have ongoing disputes about relationships; causal discovery often relies on untestable assumptions itself or only provides an equivalence class of DAGs and is commonly sensitive to hyperparameter and threshold choices. We propose an efficient, gradient-based optimization method that provides bounds for causal queries over a collection of causal graphs -- compatible with imperfect prior knowledge -- that may still be too large for exhaustive enumeration. Our bounds achieve good coverage and sharpness for causal queries such as average treatment effects in linear and non-linear synthetic settings as well as on real-world data. Our approach aims at providing an easy-to-use and widely applicable rebuttal to the valid critique of `What if your assumed DAG is wrong?'.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 46 of its 102 references that are in the map. First posted on arXiv 24 Feb 2025.
Builds on
46 of 102 references.
References · 46
- 2020Denoising Diffusion Probabilistic Modelscited by 20,851
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2022Flow Matching for Generative Modelingcited by 5,426
- 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
- 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
- 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 745
- 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 408
- 2013Causal Discovery with Continuous Additive Noise Modelscited by 252
- 2020Differentiable Causal Discovery from Interventional Datacited by 218
- 2021D'ya like DAGs? A Survey on Structure Learning and Causal Discoverycited by 209
- 2019Causality matters in medical imagingcited by 193
- 2013Causal Inference in the Presence of Latent Variables and Selection Biascited by 165
34 more
- 2021Beware of the Simulated DAG! Causal Discovery Benchmarks May Be Easy To Gamecited by 159
- 2022DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterizationcited by 135
- 2023Causal-learn: Causal Discovery in Pythoncited by 108
- 2024Causal machine learning for predicting treatment outcomescited by 73
- 2022Causal Machine Learning for Healthcare and Precision Medicinecited by 68
- 2020DAGs with No Fears: A Closer Look at Continuous Optimization for Learning Bayesian Networkscited by 66
- 2023Causal Discovery with Language Models as Imperfect Expertscited by 62
- 2020Identifying Causal-Effect Inference Failure with Uncertainty-Aware Modelscited by 57
- 2021Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confoundingcited by 50
- 2022Differentiable DAG Samplingcited by 46
- 2023Scalable Causal Discovery with Score Matchingcited by 42
- 2016The State of Applied Econometrics - Causality and Policy Evaluationcited by 41
- 2021An Automated Approach to Causal Inference in Discrete Settingscited by 39
- 2022Causal Effect Identification in Cluster DAGscited by 33
- 2020A Class of Algorithms for General Instrumental Variable Modelscited by 28
- 2019Causal Inference Under Interference And Network Uncertaintycited by 27
- 2022Stochastic Causal Programming for Bounding Treatment Effectscited by 25
- 2023Sharp Bounds for Generalized Causal Sensitivity Analysiscited by 24
- 2024Adjustment Identification Distance: A gadjid for Causal Structure Learningcited by 16
- 2023Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Modelcited by 14
- 2020Efficient adjustment sets in causal graphical models with hidden variablescited by 13
- 2021Conditional Independence Testing in Hilbert Spaces with Applications to Functional Data Analysiscited by 12
- 2022Applications of statistical causal inference in software engineeringcited by 12
- 2018The Reduced PC-Algorithm: Improved Causal Structure Learning in Large Random Networkscited by 10
- 2022Valid Inference After Causal Discoverycited by 10
- 2023Towards Bounding Causal Effects under Markov Equivalencecited by 8
- 2022Partial Identification of Dose Responses with Hidden Confounderscited by 7
- 2024Post-selection inference for causal effects after causal discoverycited by 7
- 2024A cautious approach to constraint-based causal model selectioncited by 6
- 2024Challenges and Considerations in the Evaluation of Bayesian Causal Discoverycited by 5
- 2024Estimating Causal Effects from Learned Causal Networkscited by 3
- 2022A note on efficient minimum cost adjustment sets in causal graphical modelscited by 2
- 2023Confidence in Causal Inference under Structure Uncertainty in Linear Causal Models with Equal Variancescited by 2
- 2024Dual Likelihood for Causal Inference under Structure Uncertaintycited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.