What is ML?
Machine Learning is programming by example instead of by rules.
- ✓How ML differs from traditional programming
- ✓The three broad families of ML
- ✓What 'features' and 'labels' mean
Explanation
In normal programming, a human writes the rules. In Machine Learning, you give the computer examples and it learns the rules itself.
The data has features (the inputs, e.g. square footage, location) and often a label (the answer, e.g. house price). The model learns the relationship between them.
There are three families: supervised (learn from labelled examples), unsupervised (find structure in unlabelled data), and reinforcement (learn from reward and trial-and-error).
A spam filter is ML: instead of hand-writing rules for every spam phrase, it learns from millions of emails labelled spam or not-spam.
- • Thinking ML 'figures things out' on its own — it only learns patterns present in the data you give it.
Pick a task (e.g. predicting if an email is spam) and list its likely features and label.
1. How is ML different from traditional programming?