The effect of the width parameter of the Gaussian kernel (<i>σ</i>) for a fixed value of the soft-margin constant.

<p>For large values of <i>σ</i> (A), the decision boundary is nearly linear. As <i>σ</i> decreases, the flexibility of the decision boundary increases (B). Small values of <i>σ</i> lead to overfitting (C). The figure style follows that of <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1000173#pcbi-1000173-g003" target="_blank">Figure 3</a>.