DP-100 A-Z Machine Learning using Azure ML #1

Machine Learning is the ability of the machines to learn like humans. As a kid, how did we recognise an apple? Our parents/guardians told us. Then, based on the shape and size of the object, we could recognise apples, oranges, and other fruits as well. Similar to this, machines begin to predict new data based on the prior data.

There are three ways in which machines learn—

1. Supervised Learning — This is same as a kid learning from parents based on the latter’s experiences and data fed by them to the kids. E.g., Regression (predicting a real world value like average, height, or any other metric based on the relation between different attributes passed to the model), Classification (classifying a set of data as Yes, No, Red, Black — again based on certain attributes of the data).
2. Unsupervised Learning — This form of learning happens when the machine learns on its own without human supervision.
3. Reinforcement Learning — This is same as a feedback mechanism. If we give a positive feedback to the machine to one of its output, it will try to follow similar pattern to give an output to be appreciated again. If the output is given a negative feedback, then the machine learns that the output is not up to the mark and it modifies its processing in further steps to gain maximum score via feedback.