Explaining classifier outputs with local approximations

We learn about local perturbation methods for interpretation and review SLIME.

References