Key points are not available for this paper at this time.
There has been much discussion of the right to explanation in the EU General Protection Regulation, and its existence, merits, and disadvantages. a right to explanation that opens the black box of algorithmic-making faces major legal and technical barriers. Explaining the of complex algorithmic decision-making systems and their in specific cases is a technically challenging problem. Some may offer little meaningful information to data subjects, raising around their value. Explanations of automated decisions need not on the general public understanding how algorithmic systems function. though such interpretability is of great importance and should be pursued, can, in principle, be offered without opening the black box. at explanations as a means to help a data subject act rather than understand, one could gauge the scope and content of explanations to the specific goal or action they are intended to support. From the of individuals affected by automated decision-making, we propose aims for explanations: (1) to inform and help the individual understand a particular decision was reached, (2) to provide grounds to contest the if the outcome is undesired, and (3) to understand what would need to in order to receive a desired result in the future, based on the current-making model. We assess how each of these goals finds support in the. We suggest data controllers should offer a particular type of, unconditional counterfactual explanations, to support these three. These counterfactual explanations describe the smallest change to the that can be made to obtain a desirable outcome, or to arrive at the possible world, without needing to explain the internal logic of the.
Wachter et al. (Wed,) studied this question.