Approximation of group explainers with coalition structure using Monte Carlo sampling on the product space of coalitions and features

arXiv:2303.10216v2 Announce Type: replace
Abstract: In recent years, many Machine Learning (ML) explanation techniques have been designed using ideas from cooperative game theory. These game-theoretic explainers suffer from high complexity, hindering their exact computation in practical settings. In our work, we focus on a wide class of linear game values, as well as coalitional values, for the marginal game based on a given ML model and predictor vector. By viewing these explainers as expectations over appropriate sample spaces, we design …

abstract approximation arxiv coalition complexity computation cs.lg features game game theory ideas machine machine learning math.pr practical product sampling space theory type work