How can I compare ABMs using information criterion techniques (AIC or BIC)?

[Diego Morales Bader]

I have two models in which I compare the differences with real data on eight variables. However, in addition to evaluating the error between the generated and the real data, I would like to penalize the models for the number of free parameters.

I need help to think of how the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC) could be applied to ABMs. Is this possible? Has anyone done it? What other alternatives are there?