Javascript must be enabled to continue!
A Higher Algebraic K-Theory of Causality
View through CrossRef
Causal discovery involves searching intractably large spaces. Decomposing the search space into classes of observationally equivalent causal models is a well-studied avenue to making discovery tractable. In this paper, we study the topological structure underlying causal equivalence to develop a categorical formulation of Chickering’s transformational characterization of Bayesian networks. We describe a homotopic generalization of the Meek-Chickering theorem on the connectivity structure within causal equivalence classes, and a topological representation of Greedy Equivalence Search (GES) that moves from one equivalence class of models to the next. Specifically, we define causal models as propable symmetric monoidal categories (cPROPs), a functor category CP from a coalgebraic PROP P to a symmetric monoidal category C. Such functor categories were first studied by [], who showed that they define the right adjoint of the inclusion of Cartesian categories in the larger category of all symmetric monoidal categories. cPROPs are an algebraic theory in the sense of Lawvere. CPROPs are related to previous categorical causal models, such as Markov categories and affine CDU categories, which can be viewed as defined by cPROP maps specifying the semantics of comonoidal structures corresponding to the “copy-delete" mechanisms. We develop a higher algebraic K-theory of causality by studying the classifying spaces of observationally equivalent causal cPROP models by constructing their simplicial realization through the nerve functor. We show that Meek-Chickering causal DAG equivalence generalizes to induce a homotopic equivalence across observationally equivalent cPROP functors. We present a homotopic generalization of the Meek-Chickering theorem, where covered edge reversals connecting equivalent DAGs induce natural transformations between homotopically equivalent cPROP functors and correspond to an equivalence structure on the corresponding string diagrams. We give a topological characerization of the Meek-Chickering Greedy Equivalence Search (GES) procedure. Finally, we present the Grothendieck group completion of cPROP causal models corresponding to causal DAGs using the Grayson-Quillen construction and relate the classifying space of Meek-Chickering equivalence classes to classifying spaces of an induced groupoid.
Title: A Higher Algebraic K-Theory of Causality
Description:
Causal discovery involves searching intractably large spaces.
Decomposing the search space into classes of observationally equivalent causal models is a well-studied avenue to making discovery tractable.
In this paper, we study the topological structure underlying causal equivalence to develop a categorical formulation of Chickering’s transformational characterization of Bayesian networks.
We describe a homotopic generalization of the Meek-Chickering theorem on the connectivity structure within causal equivalence classes, and a topological representation of Greedy Equivalence Search (GES) that moves from one equivalence class of models to the next.
Specifically, we define causal models as propable symmetric monoidal categories (cPROPs), a functor category CP from a coalgebraic PROP P to a symmetric monoidal category C.
Such functor categories were first studied by [], who showed that they define the right adjoint of the inclusion of Cartesian categories in the larger category of all symmetric monoidal categories.
cPROPs are an algebraic theory in the sense of Lawvere.
CPROPs are related to previous categorical causal models, such as Markov categories and affine CDU categories, which can be viewed as defined by cPROP maps specifying the semantics of comonoidal structures corresponding to the “copy-delete" mechanisms.
We develop a higher algebraic K-theory of causality by studying the classifying spaces of observationally equivalent causal cPROP models by constructing their simplicial realization through the nerve functor.
We show that Meek-Chickering causal DAG equivalence generalizes to induce a homotopic equivalence across observationally equivalent cPROP functors.
We present a homotopic generalization of the Meek-Chickering theorem, where covered edge reversals connecting equivalent DAGs induce natural transformations between homotopically equivalent cPROP functors and correspond to an equivalence structure on the corresponding string diagrams.
We give a topological characerization of the Meek-Chickering Greedy Equivalence Search (GES) procedure.
Finally, we present the Grothendieck group completion of cPROP causal models corresponding to causal DAGs using the Grayson-Quillen construction and relate the classifying space of Meek-Chickering equivalence classes to classifying spaces of an induced groupoid.
Related Results
Editorial Messages
Editorial Messages
Just as it has been continually happening in the world of mathematical sciences, the group of mathematical scientists led by (for example) Professor Eyup Cetin and his colleagues (...
Letter from the Editors
Letter from the Editors
“The present moment seems a very appropriate one to launch a new journal on Algebraic Statistics”Fabrizio Catanese, Editor of the Journal of Algebraic GeometryMany classical statis...
Macroeconomic determinants of fiscal policy in East Africa: a panel causality analysis
Macroeconomic determinants of fiscal policy in East Africa: a panel causality analysis
PurposeThis study investigates the dynamic causality linkages between fiscal deficits and selected macroeconomic indicators in a panel of five East African Community countries.Desi...
The relationship between money supply and inflation: analysis with PANELVAR approach
The relationship between money supply and inflation: analysis with PANELVAR approach
Purpose- Central banks serve as institutions responsible for executing monetary policy in countries, with the primary objective of managing the money supply and ensuring price stab...
Causality and Markovianity: Information Theoretic Measures
Causality and Markovianity: Information Theoretic Measures
Abstract
Many Information Theoretic Measures have been proposed for a quantitative assessment of causality relationships. While Gouriéroux, Monfort, and Renault (...
On algebraic systems
On algebraic systems
Abstract
The objective of this paper is to propose a generalization of algebraic closure space, namely algebraic system, and discuss its related properties. Firstly, we pro...
Analysis of Students’ Misconceptions on Solving Algebraic Contextual Problem
Analysis of Students’ Misconceptions on Solving Algebraic Contextual Problem
Students misconceptions in solving contextual algebraic problems is still often found in this educational world. This study aims to describe the forms of student’s misconceptions i...
An Analysis of Knowledge in STEM: Solving Algebraic Problems
An Analysis of Knowledge in STEM: Solving Algebraic Problems
This study was conducted to assess the students’ level of abilities in solving algebraic word problems which is a main component in Science, Technology, Engineering and Mathematics...

