AlgoPlus//machine learning / svm
Read the theory

Support Vector Machine

Separate the classes with the widest possible margin.

Margin
Support vectors
Your labelled points (x,y,class — class 1 or 0 for ±1)
Legend
Class +1
Class −1
Max-margin boundary
AI Tutor Workspace
In a nutshell
A support vector machine separates two classes by carving the widest possible empty street between them, rather than just any dividing line. The line down the middle is the boundary, and only the handful of points touching the edges of the street — the support vectors — actually decide where it goes. A wider street tends to generalize better to new data.
Ready
Press play to begin the cinematic walkthrough.
Don't just separate the classes — carve the widest empty street between them. Only the points touching the curb (support vectors) matter; the rest could move freely.
Key terms
Go deeper in the lesson
Read the full theory, intuition & complexity for Support Vector Machine.