My first day learning ML, I got through 16% of the Supervised Machine Learning: Regression and Classification by Stanford University. So far, I find that the content very interesting and the course to be very structured. My goal is to get all 3 courses of the Machine Learning Specialization done by the end of July. I find that already, I get some benefit from my A-level mathematics in high school. So far, I have tried out Jupyter notebook for the first time and have learned the differences between supervised- and unsupervised learning.

Supervised Learning

Supervised learning is the most common method of ML, and it is where you train the model based on a given set of inputs along with the expected output. That means that the model has a set of outputs to predict between, say for example, a list of housing prices and their square meter count. The square meter count would be the input and the housing price would be the expected output. The model would then make predictions for housing prices based on an input of square meters.

Unsupervised Learning

Unsupervised learning is where you train a model based on a given set of inputs but not any expected outputs. A type of unsupervised learning is clustering, where the model will group together inputs. An example of this could be a list of customers that visit a store. The inputs would then be the city they live in, and the average sum of the purchases in the store. The model will then make 'clusters' of customers that have something in common and by looking at these clusters, a person might be able to tell what cities are the most wealthy.