Towards Computing an Optimal Abstraction for Structural Causal Models
This work addresses the challenge of formalizing and learning abstractions in causal modeling, which is incremental as it builds on existing abstraction relations.
The paper tackles the problem of learning abstractions for structural causal models by defining it as an optimization of consistency and extending it with an information loss term, resulting in a concrete measure that contributes to learning new abstractions.
Working with causal models at different levels of abstraction is an important feature of science. Existing work has already considered the problem of expressing formally the relation of abstraction between causal models. In this paper, we focus on the problem of learning abstractions. We start by defining the learning problem formally in terms of the optimization of a standard measure of consistency. We then point out the limitation of this approach, and we suggest extending the objective function with a term accounting for information loss. We suggest a concrete measure of information loss, and we illustrate its contribution to learning new abstractions.