I'm trying to look for a task that predicts a discrete label first (classification), and then predicts the multiple continuous attributes of the predicted class. I found some papers about multi-output regression, but it wasn't what I wanted. Perhaps such a task exists in robot control or in video games, but I haven't found it yet. At the same time I want to train within a supervised learning framework, not reinforcement learning.
-
$\begingroup$ Hi and welcome to AI SE! Can you explain a little bit why you first need to predict a class and then predict attributees of that class? Maybe the context can also be helpful. $\endgroup$ – nbro♦ May 30 '20 at 14:28