ML – classifying taxonomy
In this section, we provide a taxonomy of ML and the basic framework of an ML system. There are various ways to define the taxonomy of ML. Broadly speaking, we can define the taxonomy as follows:
- How is the ML model trained?
- What is the learning objective?
- What to expect during model inference
- Model modality
This is a general classification of ML that is widely accepted in the community. It should be noted that there may be other taxonomies available as well. However, this should be sufficient for you to navigate all different ML algorithms.
By learning schema
Depending on how we train a model, we can place it in one of four main categories: supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning, each with its own unique set of algorithms and applications.
Figure 4.4 – Difference between supervised learning and unsupervised learning
