For the classification problem (linear/logistic regression) in machine learning, generally there is no constraint for the problem formulation. I am wondering if there are some formulations with constraints for the variables?

Similar questions and discussions