Types of ML models

As there is a selection of ML and deep learning models that address the same business problem, it is essential to understand the landscape of ML models in order to make an efficient algorithm selection. There are around 15 types of ML techniques, these being categorized into 4 categories, namely learning models, hybrid models, statistical models, and Human-In-The-Loop (HITL) models, as shown in the following matrix (where each grid square reflects one of these categories) in Figure 2.2. It is worth noting that there are other possible ways of categorizing ML models and none of them are fully complete, and as such, these categorizations will serve appropriately for some scenarios and not for others. Here is our recommended categorization with which to look at ML models:

Figure 2.2 – Types of ML models

Learning models

First, we'll take a look at two types of standard learning models, supervised learning and unsupervised learning...