I am trying to understand the class of binary classifiers that are 'probabilistic discriminative' in nature. For e.g. logistic regression is one such classifier because it is discriminative and has probabilistic interpretation. SVM is discriminative but not probabilisitic. The generalization of logistic regression to multiple classes are probabilisitic discriminative but they are not binary classifiers. Naive Bayes on the other hand is both binary and has probabilistic interpretation but is not a discriminative model.

Can someone suggest examples of binary classifiers other than logistic regression that have probabilistic discriminative interpretation. Please restrict to supervised settings only.

  • $\begingroup$ I'm sure logistic regression is not a classifier. $\endgroup$
    – SmallChess
    May 16 '17 at 6:37
  • $\begingroup$ Xgboost, neural networks $\endgroup$
    – seanv507
    Feb 10 '19 at 1:00

You can try out Linear discriminant(LDA) analysis and Quadratic discriminant(QDA) which are both probabilistic(Bayes) and discriminative at the same time.

  • $\begingroup$ (-1) LDA and QDA are generative, not discriminative $\endgroup$
    – user20160
    Feb 10 '19 at 1:25

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.