If I train a model M in a way that maximizes the penalized likelihood of M given some data, is this equivalent to maximizing its discriminative capacity? More specifically, if a major goal of the analysis is to get a highly predictive model, then should I always be training my generative model to minimize AIC rather than some other non-likelihood score/metric.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.