[sdl]

I think the paper "Explaining Explanations in AI" by Brent Mittelstadt (https://arxiv.org/abs/1811.01439v1) gives a really good overview of the different goals and approaches.