Grounded in Vapnik–Chervonenkis theory, Support Vector Machines offered maximum-margin classifiers, often with kernels mapping into high-dimensional spaces.
On text, bioinformatics, and small-scale vision benchmarks, SVMs were state of the art before deep learning. Libraries like LIBSVM democratized their use.
SVM success consolidated ML’s statistical framing versus classical symbolic AI, preparing the cross-validation and generalization culture that still rules.
