[vivin]

> The model's decision function is represented by the blue/pink background, and is clearly nonlinear. The bright red cross is the instance being explained (let's call it X). We sample instances around X, and weight them according to their proximity to X (weight here is indicated by size). We then learn a linear model (dashed line) that approximates the model well in the vicinity of X, but not necessarily globally.

So is this sort of like a local SVM classifier?

[aje403]

Not exactly, but you can "sort of" look at it that way, especially depending on the distance metric they're using to penalize distance from X, and any bias in sampling perturbed samples that are "near" X - the regression line will fit in the middle of a bunch of "support points" in the biased direction. It'd be much closer to an SVM than a regular regression line

[nerdponx]

Not an SVM. Just weighted linear regression.