1 paper · 1 filter
Jacob Trauger, Tyson Trauger, Ambuj Tewari
In this paper we will give a characterization of the learnability of forgiving 0-1 loss functions in the multiclass setting with effectively finite cardinality of the output and la…