Associate Professor in Machine Learning

Automating the Computation of Abstraction Error

less than 1 minute read

Published:

In previous posts we have explored how causal models may be related to each other at different levels of abstraction using the framework proposed by Rischel to assess their consistency and evaluate the error that may be introduced by abstraction.

In this notebook we offer a basic implementation of an exhaustive algorithm that automatically compute the abstraction between two causal models, given an abstraction between them. We show the result of the algorithm using different algorithms to enumerate all the pairs of variables over which to evaluate the abstraction error. The code is then integrated in the Abstraction class in order to provide an easily callable method to compute abstraction error.

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.