svm-pegasos-sgd-solver

IN premise — entries/2026/06/21/wiki-Support_vector_machine-chunk-6.md

Created 2026-06-21T09:55:54+00:00

Pegasos (Shalev-Shwartz et al. 2007) is a stochastic gradient descent SVM solver operating in the primal with O(1/t) convergence.

Dependents

These beliefs depend on this one:

• IN svm-solver-ecosystem-scale-dependent — SVMs have a rich ecosystem of specialized solvers whose optimal choice depends on problem geometry — SMO decomposes into 2D subproblems for general QP, Pegasos uses stochastic gradient descent in the primal for large-scale problems, LIBLINEAR achieves linear-time iterations for high-dimensional sparse data, and the crossover point between sub-gradient and coordinate descent depends on whether n or d dominates.