Causal structure learning: A combinatorial perspective

C Squires, C Uhler - Foundations of Computational Mathematics, 2023 - Springer
In this review, we discuss approaches for learning causal structure from data, also called
causal discovery. In particular, we focus on approaches for learning directed acyclic graphs …

Consistency guarantees for greedy permutation-based causal inference algorithms

L Solus, Y Wang, C Uhler - Biometrika, 2021 - academic.oup.com
Directed acyclic graphical models are widely used to represent complex causal systems.
Since the basic task of learning such a model from data is NP-hard, a standard approach is …

A Fixed-Parameter Tractable Algorithm for Counting Markov Equivalence Classes with the Same Skeleton

VS Sharma - Proceedings of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
Causal DAGs (also known as Bayesian networks) are a popular tool for encoding
conditional dependencies between random variables. In a causal DAG, the random …

Markov equivalence of max-linear Bayesian networks

C Améndola, B Hollering… - Uncertainty in Artificial …, 2021 - proceedings.mlr.press
Max-linear Bayesian networks have emerged as highly applicable models for causal
inference from extreme value data. However, conditional independence (CI) for max-linear …

An efficient algorithm for counting Markov equivalent DAGs

R Ganian, T Hamm, T Talvitie - Artificial Intelligence, 2022 - Elsevier
We consider the problem of counting the number of DAGs which are Markov equivalent, ie,
which encode the same conditional independencies between random variables. The …

On the edges of characteristic imset polytopes

S Linusson, P Restadh, L Solus - arxiv preprint arxiv:2209.07579, 2022 - arxiv.org
The edges of the characteristic imset polytope, $\operatorname {CIM} _p $, were recently
shown to have strong connections to causal discovery as many algorithms could be …

Lazyiter: a fast algorithm for counting Markov equivalent DAGs and designing experiments

A AhmadiTeshnizi, S Salehkaleybar… - … on Machine Learning, 2020 - proceedings.mlr.press
The causal relationships among a set of random variables are commonly represented by a
Directed Acyclic Graph (DAG), where there is a directed edge from variable $ X $ to variable …

On the number and size of Markov equivalence classes of random directed acyclic graphs

D Schmid, A Sly - arxiv preprint arxiv:2209.04395, 2022 - arxiv.org
In causal inference on directed acyclic graphs, the orientation of edges is in general only
recovered up to Markov equivalence classes. We study Markov equivalence classes of …

Size of interventional Markov equivalence classes in random DAG models

D Katz, K Shanmugam, C Squires… - The 22nd International …, 2019 - proceedings.mlr.press
Directed acyclic graph (DAG) models are popular for capturing causal relationships. From
observational and interventional data, a DAG model can only be determined up to its\emph …

Merging joint distributions via causal model classes with low VC dimension

D Janzing - arxiv preprint arxiv:1804.03206, 2018 - arxiv.org
If $ X, Y, Z $ denote sets of random variables, two different data sources may contain
samples from $ P_ {X, Y} $ and $ P_ {Y, Z} $, respectively. We argue that causal inference …