Algorithm 1. LIME

Require: Classifier f, Number of samples N Require: Instance x, and its interpretable version $x^{\prime }$ Require: Similarity kernel πx, Length of explanation K $Z$ ← {} for i ∈ {1, 2, 3,..., N} do   $z_i^{\prime }$ ← Sample around ($x^{\prime }$)   $Z$ ← $Z$ ∪ ($z_i^{\prime }$, f$(z^i$), ${\mathsf{\pi }}_x$ ($z_i$)) end for w ← K-Lasso(Z, K)     ▷ with $z_i^{\prime }$ as features, f(z) as target return w