What is a “surrogate model”?

Question

While reading about model explainability and model accountability, the term surrogate model keeps appearing. I had an idea about what it is but it does not seem to make sense anymore:

• What is a surrogate models?

• Why are surrogate models good for explainability?

Answer 1

A "surrogate" is just a stand-in or proxy. In data science the word "surrogate" is used in more than one way (Bayesian hyperparameter optimization comes to mind). For interpretability, it seems to be used mostly to mean a more-interpretable model (maybe linear/logistic regression) that is trained to approximate the main, usually black-box, model (maybe boosted trees or a neural network).

See e.g. https://christophm.github.io/interpretable-ml-book/global.html. (The author refers to LIME as a "local surrogate" method and differentiates the "global surrogate" method, but I'm used to hearing just "surrogate" as a global one.)

Answer 2

A surrogate model is an approximation model for a given function.

The original function is generally a black box function that we can sample from and based on the samples we can optimize our surrogate model to approximate the behaviour of the original function.

A surrogate model can be a neural network, an ensemble method, a gaussian process which can be interpreted and used later for explainability.