Classification

Logistic Regression

Unfortunately, there is no closed-form solution for maximizing the log-likelihood (or minimizing the inverse, the logistic cost function); at least it has not been found, yet.

Neural Network Classification

Log, Exp, Cross Entropy

SVM

Two class SVM

SVM Loss

Multi-class SVM vs. Cross Entropy

How to solve SVM

Decision Tree

Random Forest

GBDT

Gradient Boosted Decision Trees - Module 4: Supervised Machine Learning - Part 2 | CourseraCoursera

Bayesian Network

Nearest Neighbor Classification