multinomial logistic regression
(30 minutes to learn)
Summary
Multinomial logistic regression is a generalization of logistic regression to the case where there are more than two categories.
Context
This concept has the prerequisites:
- logistic regression
- maximum likelihood (Multinomial logistic regression can be fit using maximum likelihood.)
- optimization problems (Multinomial logistic regression is formulated as an optimization problem.)
- gradient descent (Multinomial logistic regression can be fit using gradient descent.)
Goals
- Know what the multinomial logistic regression model is
- Derive the gradient descent update rule
Core resources (read/watch one of the following)
-Paid-
→ Pattern Recognition and Machine Learning
A textbook for a graduate machine learning course, with a focus on Bayesian methods.
Location:
Section 4.3.4, "Multiclass logistic regression," pages 209-210
Supplemental resources (the following are optional, but you may find them useful)
-Paid-
→ Machine Learning: a Probabilistic Perspective
A very comprehensive graudate-level machine learning textbook.
Location:
Section 8.3.7, "Multi-class logistic regression," pages 252-254
See also
- Here are some other commonly used multiway classification algorithms: Multinomial logistic regression is a discriminative model .
- Multinomial logistic regression is a kind of generalized linear model .