MAchine learning models

 Traditional And Very Useful Machine Learning Models

All the data used in this chapter is labeled with ground truths, and the goal of our models is to predict the labels of new (unseen) and unlabeled data. This is supervised learning.

In the next few sections, we fit training functions into our labeled data using the following popular machine learning models:

• Linear Regression: Predict a numerical value.
• Logistic Regression: Classify into two classes (binary classification).
• Softmax Regression: Classify into multiple classes.
• Support Vector Machines: Classify into two classes, or regression (predict a numerical value).
• Decision Trees: Classify into any number of classes, or regression (predict a numerical value).
• Random Forests: Classify into any number of classes, or regression (predict a numerical value).
• Ensembles of models: Bundle up the results of many models, by averaging the prediction values, voting for the most popular class, or some other bundling mechanism.

Numerical Solutions vs. Analytical Solutions

It is important to be aware of the difference between numerical solutions and analytical solutions of mathematical problems. A mathematical problem can be anything, such as:

• Find the minimizer of some function.
• Find the best way to go from destination A to destination B, with a constrained budget.
• Find the best way to design and query a data warehouse.
• Find the solution of a mathematical equation (where a left hand side with math stuff equals a right hand side with math stuff). These equations could be algebraic equations, ordinary differential equations, partial differential equations, integro-differential equations, systems of equations, or any sort of mathematical equations. Their solutions could be static or evolving in time. They could model anything from the physical, biological, socioeconomical or natural worlds.

Here is the vocabulary:

Numerical: has to do with numbers.

Analytical: has to do with analysis.