Algorithmic approaches to training Support Vector Machines: a survey

: Support Vector Machines (SVMs) have become an increasingly popular tool for machine learning tasks involving classi cation, regression or novelty detection. They exhibit good generalisation performance on many reallife datasets and the approach is well-motivated theoretically. Training involves optimisation of a convex cost function, there are relatively few free parameters to adjust and the architecture does not have to be found by experimentation. In this tutorial we survey methods for training SVMs including model selection strategies for determining the free parameters and new techniques for active selection of training examples.

• Type: Conference
• Year: 2000
• Where: ESANN
• Authors: Colin Campbell
• Keywords: ESANN 2000, ESANN 2007, Free Parameters, Tasks Involving Classi, Training Involves Optimisation