Hill climbing and loss functions

In the last section, we got comfortable with the idea of supervised machine learning. Now, we will learn how exactly a machine learns underneath the hood. This section is going to examine a common optimization technique used by many machine learning algorithms, called hill climbing. It is predicated on the fact that each problem has an ideal state and a way to measure how close or how far we are from that. It is important to note that not all machine learning algorithms use this approach.