Associate Professor in Machine Learning

Category-Theoretical Evaluation of Abstraction between Causal Models

less than 1 minute read

Published:

Causal models offer a rigorous formalism to express causal relations between variables of interest. Causal systems may be represented at different levels of granularity or abstraction; think, for example, to microscopic and macroscopic descriptions of thermodynamics systems. Reasoning about the relationship between causal models at different levels of abstraction is a non trivial problem.

In this notebook we provide a tutorial introduction to the categorical framework proposed by Rischel which relies on a category theory formalization to evaluate and measure the approximation (or error) introduced by abstractions between causal models. We illustrate ideas and concepts running simulations inspired by the examples provided by the author, and we then discuss the meaning and the implications of the results.

Notebook here

You May Also Enjoy

Reviewing Consistencies

less than 1 minute read

Published:

In previous posts we have discussed theoretical and conceptual properties of abstractions between causal models, while at the same time implementing code for checking our statements and for running simulations.

Scenarios of Confounding

less than 1 minute read

Published:

Causality and causal inference deal with expressing and reasoning about relationships of cause and effects, and structural causal model provide a rigorous formalism to assess causality.