AI glossary

The SchoolWhool AI glossary defines 853 terms from machine learning and the subjects under it, each in one line, the way the lessons use them. Every term links to the lesson that teaches it, so you can go from a definition to the worked example and the code.

The terms are the key terms of 121 published lessons in the Python, Data Tools, Maths, Statistics and Machine Learning courses. A term used in several lessons links to each of them. The glossary grows as lessons are published.

Start here

12 terms the rest of the glossary builds on.

machine learning
Choosing a program's parameters from examples, then using it on new examples.
regression
Supervised learning where the answer is a quantity, so being close counts.
classification
Supervised learning where the answer is one of a set of categories.
label
The answer a program should give for an example.
overfitting
Fitting the training rows much better than new rows.
cross-validation
Estimating a model's score by fitting and validating on several splits of the training rows.
gradient descent
Repeatedly stepping the parameters against the gradient of the loss.
learning rate
The hyperparameter that scales each step, η.
derivative
the slope of the tangent, the limit of Δy/Δx
vector
an ordered list of numbers, written bold
mean
The sum of observed values divided by their count.
residual
actual value minus predicted value

Browse all 853 terms, A to Z

Looking for a whole subject? See all courses