https://abaj.ai/tags/max-margin/Recent content in Max-Margin on Aayush Bajaj's Augmenting InfrastructureHugoen© 2026 Aayush BajajThu, 09 Jul 2026 21:02:19 +1000https://abaj.ai/wiki/ml/supervised/classification/svm/Thu, 09 Jul 2026 21:02:56 +1000https://abaj.ai/wiki/ml/supervised/classification/svm/<p>a linearly separable dataset admits infinitely many separating hyperplanes, and the <a href="https://abaj.ai/wiki/ml/supervised/classification/perceptron/" >perceptron</a> will happily hand you whichever one it trips over first.<span class="margin-note" data-note="the perceptron&#39;s answer depends on initialisation and the order you feed it data. the svm&#39;s answer is unique."> <span class="margin-note-indicator">𐃏</span> </span> the support vector machine asks a better question: of all the hyperplanes that separate the data, which one is <em>farthest from everybody</em>? the answer — the maximum-margin hyperplane — is determined by a handful of boundary points (the support vectors), drops out of a beautiful convex dual, and generalises via the kernel trick from lines to nearly anything.</p>