报告环节主持

祁   琦 中国人民大学高瓴人工智能学院长聘副教授  

09:30-

10:00

主题报告一

报告题目：Eliciting Information via Crowd Intelligence

报告摘要：Many online social networking platforms leverage crowdsourcing to enhance the user experiences. These platforms seek to incentivize workers to exert efforts to complete tasks (e.g., evaluating the quality of online posts and articles) and truthfully report their findings. Such problems can be very challenging to solve when it is too expensive or impossible to verify the ground truth. We will introduce our recent results on designing effective incentive mechanisms to induce truthful information revelation from crowds, through careful output agreement and truth detection mechanism design.

主讲人:

黄建伟  香港中文大学（深圳）校长讲座教授\理工学院副院长

10:30-

11:00

主题报告三

报告题目：Truthful Cake Sharing

报告摘要：The classic cake cutting problem concerns the fair allocation of a heterogeneous resource among interested agents. In this talk, we discuss a public goods variant of the problem, where instead of competing with one another for the cake, the agents all share the same subset of the cake which must be chosen subject to a length constraint. We focus on the design of truthful and fair mechanisms in the presence of strategic agents who have piecewise uniform utilities over the cake. We show that the leximin solution is truthful and moreover maximizes an egalitarian welfare measure among all truthful and position oblivious mechanisms.

主讲人：

贝小辉  新加坡南洋理工大学副教授

11:00-

11:30

主题报告四

报告题目：Tight Incentive Analysis on Sybil Attacks to Market Equilibrium of Resource Exchange over General Networks

报告摘要：The Internet-scale peer-to-peer (P2P) systems usually build their success on distributed protocols. For example, the well-known BitTorrent network for resource exchange is based on the proportional response protocol, where each participant exchanges its resources with its neighbors in proportion to what it has received in the previous round. The dynamics of such a protocol has been proved to converge to a market equilibrium. On the other hand, it requires thorough incentive analysis to show the robustness of such protocol, as the distributed agents may strategically manipulate the system once they are able to benefit. Recent studies have developed strategy-proofness results of the proportional response protocol against agent deviations in the forms of weight cheating and edge deleting. However, the protocol is not truthful against Sybil attacks, under which an agent may create several fictitious identities and control these fictitious identities to exchange resource with others. In this paper, we apply the concept of incentive ratio to measure how much the utility of a strategic agent in a market equilibrium can be improved by playing Sybil attacks. Similar to approximation ratio, price of anarchy and competitive ratio, a smaller incentive ratio means how close our solution is for an agent not to manipulate the system. We prove a tight incentive ratio of two for any agent launching Sybil attacks over general networks. The tight incentive ratio of two closes an open problem modeling the successful tit-for-tat protocol for Internet resource exchanging, and also presents a complete picture in this line of theoretical studies with real applications.

主讲人：

程郁琨  苏州科技大学教授

午间休息

报告环节主持

王子贺 中国人民大学高瓴人工智能学院准聘助理教授  

12:50-13:20

主题报告五

报告题目：房产税改革的经济与福利效应—基于异质性个体模型的量化研究

报告摘要：如何理解房产税在宏观经济运行中的作用和对微观家庭的异质性影响，对于中国的房产税改革非常重要。本文构建了一个包含异质性家户的世代交叠模型，量化分析了房产税改革对异质性微观个体决策和宏观经济运行的影响。校准的基准模型可以同时匹配中国的宏观经济指标和家户的微观行为。研究表明，房产税改革将优化居民的资产结构，刺激消费，降低房价并显著的提振实体经济。同时，房产税改革带来了显著的福利改善，低收入群体福利改善最明显，而高收入群体福利改善最低，征收房产税具有再分配效应。当存在征收面积减免时，居民更倾向于购买临界面积的房产，穷人更容易买房的同时伴随着消费和福利的下降，房产税的再分配效应减弱。在转移路径上，征收房产税短期内会带来房价的下降和租金的急剧上升，居民自有房率在改革初期下降明显。

主讲人：

李  博  北京大学经济学院经济学系助理教授

13:20-13:50

主题报告六

报告题目：Eliciting Information without Verification

报告摘要：When we use the wisdom of the crowds, we usually rank the answers according to their popularity, especially when we cannot verify the answers. However, this can be very dangerous when the majority make systematic mistakes. A fundamental question arises: can we build a hierarchy among the answers without any prior where the higher-ranking answers, which may not be supported by the majority, are from more sophisticated people? To address the question, we propose 1) a novel model to describe people's thinking hierarchy; 2) two algorithms to learn the thinking hierarchy without any prior; 3) a novel open-response based crowdsourcing approach based on the above theoretic framework. In addition to theoretic justifications, we conduct four empirical crowdsourcing studies and show that a) the accuracy of the top-ranking answers learned by our approach is much higher than that of plurality voting (In one question, the plurality answer is supported by 74 respondents but the correct answer is only supported by 3 respondents. Our approach ranks the correct answer the highest without any prior); b) our model has a high goodness-of-fit, especially for the questions where our top-ranking answer is correct.

主讲人：

孔雨晴   北京大学前沿计算研究中心助理教授

13:50-14:20

主题报告七

报告题目：Stable Market Segmentation against Price Discrimination

报告摘要：In this paper, we analyze stable market segmentation where consumers are mobile among different markets. Free circulation is legally mandated by data protection regulations, which endogenizes market segmentation. We show that in every stable market segmentation, the producer surplus remains at the uniform monopoly level, and the consumer surplus takes a value between the buyer-optimal level and the uniform monopoly level. No consumer is worse off than the uniform monopoly: Anonymous consumers are charged with the uniform monopoly price, while other consumers trade their privacy for a discount. Therefore, our results justify the Pareto optimum of price discrimination and reveal welfare implications of current regulations.

主讲人：

邝仲弘   中国人民大学经济学院讲师

茶歇（10分钟）

报告环节主持

沈蔚然 中国人民大学高瓴人工智能学院准聘助理教授  

14:30-15:00

主题报告八

报告题目：On Existence of Truthful Fair Cake Cutting Mechanisms

报告摘要：We study the fair division problem on divisible heterogeneous resources (the cake cutting problem) with strategic agents, where each agent can manipulate his/her private valuation in order to receive a better allocation. A (direct-revelation) mechanism takes agents’ reported valuations as input, and outputs an allocation that satisfies a given fairness requirement. A natural and fundamental open problem, first raised by Chen et al. and subsequently raised by Procaccia, Aziz and Ye, Brânzei and Miltersen, Menon and Larson, Bei et al., etc., is whether there exists a deterministic, truthful and envy-free (or even proportional) cake cutting mechanism. In this paper, we resolve this open problem by proving that there does not exits a deterministic, truthful and proportional cake cutting mechanism, even in the special case where all of the followings hold:

1) there are only two agents;

2) each agent’s valuation is a piecewise-constant function;

3) each agent is hungry: each agent has a strictly positive value on any part of the cake.

The impossibility result extends to the case where the mechanism is allowed to leave some part of the cake unallocated.

To circumvent this impossibility result, we aim to design mechanisms that possess certain degree of truthfulness. Motivated by the kind of truthfulness possessed by the classical I-cut-you-choose protocol, we define a weaker notion of truthfulness: the risk-averse truthfulness. We show that the well-known moving-knife procedure and Even-Paz algorithm do not have this truthful property. We propose a mechanism that is risk-averse truthful and envy-free, and a mechanism that is risk-averse truthful and proportional that always outputs allocations with connected pieces.

This paper has been accepted by EC’22.

主讲人：

陶表帅 上海交通大学电子信息与电气工程学院助理教授

15:00-15:30

主题报告九

报告题目：A Best Cost-Sharing Rule for Selfish Bin Packing

报告摘要：In selfish bin packing, each item is regarded as a player, who aims to minimize the cost-share by choosing a bin it can fit in. To have a least number of bins used, cost-sharing rules play an important role. The currently best known cost sharing rule has a lower bound on PoA larger than 1.45, while a general lower bound 4/3 on PoA applies to any cost-sharing rule under which no items have incentive unilaterally moving to an empty bin. We propose a novel and simple rule with a PoA matching the lower bound, thus completely resolving this game. The new rule always admits a Nash equilibrium and its PoS is one. Furthermore, the well-known bin packing algorithm BFD (Best-Fit Decreasing) is shown to achieve a strong equilibrium, implying that a stable packing with an asymptotic approximation ratio of 11/9 can be produced in polynomial time.

主讲人：

王长军  中科院数学与系统科学研究院副研究员

15:30-16:00

主题报告十

报告题目：Mechanism Design Powered by Social Interactions

报告摘要：Mechanism design has traditionally assumed that the participants are fixed and independent. However, in reality, the participants are well-connected (e.g., via their social networks) and we can utilize their connections to power the design. One interesting trend is to incentivize the existing participants of a game to use their connections to invite new participants. This helps to form larger games in auctions, coalitional games, matching etc., which cannot be done with the traditional solutions. The new challenge is that the participants are competitors and they would not invite each other by default. This challenge seems more complex than some of the challenges the community investigated before. For example, in auctions, solving it may require revenue monotonicity and false-name-proofness, which were proved impossible to achieve in most cases. In matching, this cannot get along with standard optimality and stability. Hence, we believe this new trend is worth rich investigations and will stimulate many interesting applications, especially under decentralized systems with blockchain.

主讲人：

赵登吉   上海科技大学准聘制助理教授

16:00-

16:30

闭门研讨会（会议链接另行分享）

主持

祁   琦 中国人民大学高瓴人工智能学院长聘副教授