Chapter 12: Multiclass Classification
This chapter treats the multiclass case of classification. Tasks with more than two classes preclude the application of some techniques studied in the binary scenario and require an adaptation of loss functions.
-
Chapter 12.01: Multiclass Classification and Losses
In this section, we introduce the basic concepts in multiclass (MC) classification and important MC losses: MC 0-1 loss, MC brier score, and MC logarithmic loss.
-
Chapter 12.02: Softmax Regression
In this section, we introduce softmax regression as a generalization of logistic regression.
-
Chapter 12.03: One-vs-One and One-vs-Rest
It is sometimes advisable to address a multiclass problem as a set of binary ones. We discuss two ways to reduce a multiclass problem to multiple binary classification problems: one-vs-one and one-vs-rest.
-
Chapter 12.04: Designing Codebooks and ECOC
In this section, we introduce codebooks as a general concept for multiclass-to- binary reduction.