The Limits of Multi-task Peer Prediction

The Limits of Multi-task Peer Prediction
复制标题

多任务同行预测的局限性

DOI:
10.1145/3465456.3467642
复制
发表时间:
2021
期刊:
EC '21: Proceedings of the 22nd ACM Conference on Economics and Computation
影响因子:
--
通讯作者:
Chen, Yiling
Chen, Yiling
中科院分区:
--
文献类型:
--
作者:
Zheng, Shuran;Yu, Fang-Yi;Chen, Yiling

文献摘要

参考文献

被引文献

相似文献

多任务同伴预测的最新进展极大地扩展了我们对多任务同伴预测机制的认识。在不同的环境中提出了不同的机制来引出不同类型的信息。但是,对于多任务同伴预测问题何时会存在理想的机制,我们仍然缺乏理解。在这项工作中,我们研究了多任务同伴预测问题的适宜性。我们认为设计师对潜在的信息结构有一定的了解,并希望从一组参与者中引出某些信息。我们的目标是根据问题的原语推断出拥有理想机制的可能性。我们的贡献是双重的。首先,假设设计者只使用评分机制,我们提供了可引出的多任务同伴预测问题的特征。评分机制是对参与者对不同任务的报告分别进行奖励的机制。在单任务设置中,表征使用基于功率图表征的几何方法。对于一般机制,我们也给出了多任务问题可引出的必要条件。其次,我们考虑的情况是,当设计师的目标是在参与者关于世界状态的后验中引出一些线性属性。我们首先表明,在某些情况下,设计师基本上只能引出后验本身。然后,我们研究当设计师的目标是引出参与者的后验时的情况。我们给出了后验可引出的必要条件。这个条件意味着Kong和Schoenebeck提出的机制已经是我们在他们的设定中所能期望的最好的机制,因为他们的机制可以解决任何可能引起的问题实例。
Recent advances in multi-task peer prediction have greatly expanded our knowledge about the power of multi-task peer prediction mechanisms. Various mechanisms have been proposed in different settings to elicit different types of information. But we still lack understanding about when desirable mechanisms will exist for a multi-task peer prediction problem. In this work, we study the elicitability of multi-task peer prediction problems. We consider a designer who has certain knowledge about the underlying information structure and wants to elicit certain information from a group of participants. Our goal is to infer the possibility of having a desirable mechanism based on the primitives of the problem. Our contribution is twofold. First, we provide a characterization of the elicitable multi-task peer prediction problems, assuming that the designer only uses scoring mechanisms. Scoring mechanisms are the mechanisms that reward participants' reports for different tasks separately. The characterization uses a geometric approach based on the power diagram characterization in the single-task setting. For general mechanisms, we also give a necessary condition for a multi-task problem to be elicitable. Second, we consider the case when the designer aims to elicit some properties that are linear in the participant's posterior about the state of the world. We first show that in some cases, the designer basically can only elicit the posterior itself. We then look into the case when the designer aims to elicit the participants' posteriors. We give a necessary condition for the posterior to be elicitable. This condition implies that the mechanisms proposed by Kong and Schoenebeck are already the best we can hope for in their setting, in the sense that their mechanisms can solve any problem instance that can possibly be elicitable.
DOI: 10.1145/1386790.1386813
发表时间: 2008-07
期刊: --
影响因子: --
作者:
Nicolas S. Lambert;David M. Pennock;Y. Shoham
通讯作者: Nicolas S. Lambert;David M. Pennock;Y. Shoham
任务数量恒定的显性真实多任务同行预测
DOI: --
发表时间: 2019
期刊: ACM-SIAM Symposium on Discrete Algorithms
影响因子: --
作者:
Yuqing Kong
通讯作者: Yuqing Kong
两块岩石中的水:最大化互信息
DOI: --
发表时间: 2018
期刊: ACM Conference on Economics and Computation
影响因子: --
作者:
Yuqing Kong;G. Schoenebeck
通讯作者: G. Schoenebeck
数据科学的博弈论:获取真实信息
DOI: --
发表时间: 2017
期刊: Game Theory for Data Science
影响因子: --
作者:
B. Faltings;Goran Radanovic
通讯作者: Goran Radanovic
三个异质代理回答一个问题的两种强真实机制
DOI: --
发表时间: 2022
期刊: Workshop on Internet and Network Economics
影响因子: --
作者:
G. Schoenebeck;Fang
通讯作者: Fang