Algorithmic Mechanism Design With Investment
研究了使用近似算法的真实机制对投资激励的影响，发现分配效率高的算法可能完全无法保证投资激励，并提出了无确认负外部性的快速近似算法，同时保证高分配效率和高投资激励。
We study the investment incentives created by truthful mechanisms that allocate resources using approximation algorithms. Some approximation algorithms guarantee nearly 100% of the optimal welfare in the allocation problem but guarantee nothing when accounting for investment incentives. An algorithm's allocative and investment guarantees coincide if and only if its confirming negative externalities are sufficiently small. We introduce fast approximation algorithms for the knapsack problem that have no confirming negative externalities and guarantees close to 100% for both allocation and investment.