Towards Attributions of Input Variables in a Coalition

This paper aims to develop a new attribution method to explain the conflict between individual variables' attributions and their coalition's attribution from a fully new perspective. First, we find that the Shapley value can be reformulated as the allocation of Harsanyi interactions encoded by the AI model. Second, based the re-allocation of interactions, we extend the Shapley value to the attribution of coalitions. Third, we derive the fundamental mechanism behind the conflict. This conflict comes from the interaction containing partial variables in their coalition.