Causal Chambers: Real Physical Systems as a Testbed for Validating AI Algorithms
The Causal Chambers provide real-world datasets from well-understood physical systems to serve as a testbed for validating a variety of AI, machine learning, and statistical algorithms.
Confidence Sets for Causal Orderings in Identifiable Structural Equation Models with Additive Errors
This research paper introduces a novel method for constructing confidence sets of causal orderings in identifiable structural equation models with additive errors, addressing the challenge of quantifying uncertainty in causal discovery.
Causal Discovery with Parent Scores for Mixed Linear and Nonlinear Relations: Introducing the CaPS Algorithm
This paper introduces CaPS, a novel ordering-based causal discovery algorithm that effectively handles datasets containing both linear and nonlinear causal relationships, outperforming existing methods restricted to a single type of relation.
A New Method for Causal Discovery in the Presence of Heteroscedastic Noise Using Skewness of Score Functions
This paper introduces a novel method, SkewScore, for identifying causal relationships in data with heteroscedastic noise by leveraging the skewness of the score function, offering a computationally efficient alternative to existing approaches.
DiffIntersort: A Differentiable and Scalable Approach to Causal Order Discovery from Interventional Data
This paper introduces DiffIntersort, a novel method that improves upon the existing Intersort algorithm for identifying causal relationships between variables using interventional data. DiffIntersort overcomes scalability limitations in Intersort, enabling its application to larger datasets and integration into gradient-based machine learning workflows.
LoSAM: A Top-Down Global Discovery Approach for Causal Discovery in Additive Noise Models with Unmeasured Confounders
LoSAM is a novel algorithm that efficiently identifies causal relationships in data generated by complex systems, even when some influencing factors are hidden, by leveraging local causal structures and advanced statistical techniques.
MissNODAG: A Differentiable Framework for Learning Cyclic Causal Graphs from Incomplete Data, Including MNAR Mechanisms
MissNODAG is a novel framework that effectively learns cyclic causal relationships from incomplete data, addressing limitations of existing methods by handling both MNAR missingness and feedback loops in systems.
A Novel Sequential Method for Causal Order Discovery in Monotonic Structural Causal Models Using Jacobian of TMI Maps
This paper introduces a new sequential method for discovering the causal order of variables in Monotonic Structural Causal Models (SCMs) by iteratively identifying the root variable using the Jacobian of Triangular Monotonic Increasing (TMI) maps, eliminating the need for sparsity assumptions and outperforming existing methods based on Jacobian sparsity maximization.
Sample Efficient Bayesian Learning of Causal Graphs from Interventions: A Novel Algorithm for Causal Discovery with Limited Data
This research paper introduces a novel Bayesian algorithm designed to efficiently learn causal graphs from a limited number of interventional samples, addressing the real-world challenge of costly interventions in causal discovery.
Causal Modeling in Multi-Context Systems: Identifying Context-Specific Causal Relationships from Observational Data
This paper proposes a novel framework for understanding and distinguishing between different types of context-specific changes in causal relationships within multi-context systems, addressing the challenge of identifying causal structures when observational support varies across contexts.