Understanding the basics of machine learning

Before we get started with using machine learning for time series, let’s spend some time establishing what machine learning is and setting up a framework to demonstrate what it does (if you are already very comfortable with machine learning, feel free to skip ahead, or just stay with us and refresh the concepts). In 1959, Arthur Samuel defined machine learning as a “field of study that gives computers the ability to learn without being explicitly programmed.” Traditionally, programming has been a paradigm under which we know a set of rules/logic to perform an action, and that action is performed on the given data to get the output that we want. But, machine learning flipped this on its head. In machine learning, we start with data and the output, and we ask the computer to tell us about the rules with which the desired output can be achieved from the data:

Figure 5.1 – Traditional programming...