A Characterization of Online Multiclass Learnability
share
We consider the problem of online multiclass learning when the number of labels is unbounded. We show that the Multiclass Littlestone dimension, first introduced in <cit.>, continues to characterize online learnability in this setting. Our result complements the recent work by <cit.> who give a characterization of batch multiclass learnability when the label space is unbounded.
READ FULL TEXT