Motivated by the fact that the worth of a coalition may depend on the order in which agents arrive, Nowak and Radzik (1994) introduced cooperative games with generalized characteristic functions, where the worth function is defined on sequences of agents rather than sets. We call such games temporal cooperative games (TCGs). This order sensitivity requires re-examining axioms for reward sharing. Prior work proposed several axioms; however, the resulting solution concepts are order-oblivious and closely tied to the Shapley value. In contrast, we focus on sequential solution concepts that explicitly depend on the realized order. We study reward-sharing mechanisms satisfying incentive for optimal arrival (I4OA), which promotes orders that maximize total worth; online individual rationality (OIR), which ensures later arrivals do not harm existing agents; and sequential efficiency (SE), which requires that the worth of any sequence is fully distributed among its agents. These axioms are intrinsic to TCGs, and we characterize a class of reward-sharing mechanisms uniquely determined by them. The classical Shapley value does not directly extend to this setting. We therefore construct Shapley analogs in two worlds: a sequential world, where rewards are defined for each sequence-agent pair, and an extended world, where rewards are defined per agent, consistent with Nowak and Radzik (1994). In both cases, efficiency, additivity, and null player uniquely characterize the Shapley analogs. However, these analogs are disjoint from the class of solutions satisfying the sequential axioms, even for convex and simple TCGs. Our results reveal a fundamental tension: when order matters, solution concepts satisfying natural sequential incentives must differ structurally from Shapley-based allocations, motivating new axiomatic foundations for sequential reward sharing.
翻译:暂无翻译