论文标题
防止策略和嫉妒的房屋分配机制
Strategy-proof and Envy-free Mechanisms for House Allocation
论文作者
论文摘要
当代理对物体严格偏好时，我们考虑将不可分割的对象分配给代理的问题。效率，公平性和分配机制的激励措施之间的竞争概念之间存在固有的权衡。因此，自然而然地考虑以最强的概念满足这三个特性中的两种机制，同时试图改善第三维。在本文中，我们受到以下问题的激励：是否有比平等分裂更有效的防策略和嫉妒的随机分配机制？ 我们在本文中的贡献是双重的。首先，我们进一步探讨了防止策略机制类别中效率和嫉妒性之间的不相容性。我们定义了一种新的效率概念，比前部件效率弱，并证明即使在这种非常弱的意义上，任何防止策略和嫉妒的机制也必须牺牲效率。接下来，我们介绍了一种称为成对交换机制的新机制家族，并令人惊讶地观察到，防止策略是等同于此类中的嫉妒性。我们表征了这个家族中所有中性和防止策略的机制（因此，也是嫉妒的）机制的集合，并表明它们承认了非常简单的线性表示。
We consider the problem of allocating indivisible objects to agents when agents have strict preferences over objects. There are inherent trade-offs between competing notions of efficiency, fairness and incentives in assignment mechanisms. It is, therefore, natural to consider mechanisms that satisfy two of these three properties in their strongest notions, while trying to improve on the third dimension. In this paper, we are motivated by the following question: Is there a strategy-proof and envy-free random assignment mechanism more efficient than equal division? Our contributions in this paper are twofold. First, we further explore the incompatibility between efficiency and envy-freeness in the class of strategy-proof mechanisms. We define a new notion of efficiency that is weaker than ex-post efficiency and prove that any strategy-proof and envy-free mechanism must sacrifice efficiency even in this very weak sense. Next, we introduce a new family of mechanisms called Pairwise Exchange mechanisms and make the surprising observation that strategy-proofness is equivalent to envy-freeness within this class. We characterize the set of all neutral and strategy-proof (and hence, also envy-free) mechanisms in this family and show that they admit a very simple linear representation.
