~/blog/tutorials/machine-learning

Svm

Jun 26, 20269 min read

SVM: Hard Margin and Soft Margin

Logistic regression stops when the loss is low. Given a linearly separable dataset, infinitely many hyperplanes achieve zero training loss — logistic regression…

Tutorial

Jun 26, 20269 min read

SVM: Math Intuition and Cost Function

The previous post defined the SVM objective — maximize the margin — and showed that maximizing $2/\|w\|$ is equivalent to minimizing $\|w\|^2/2$. What it didn't…

Tutorial

Jun 26, 20269 min read

SVM Kernels

The linear SVM finds the maximum-margin hyperplane in the original feature space. When classes aren't linearly separable — like the XOR pattern — no hyperplane…

Tutorial

Jun 26, 20268 min read

SVC and SVR: Full Implementation

Theory and kernel math are complete. This post runs SVC end-to-end on Breast Cancer classification and SVR on California Housing regression. Every number is ver…

Tutorial

Jun 27, 202613 min read

The Kernel Trick

The kernel trick lets you train an SVM in a space with millions — or infinitely many — dimensions without ever computing a single coordinate in that space. The…