通过有限的真值访问激励评估：同行预测使事情变得更糟
Incentivizing Evaluation via Limited Access to Ground Truth: Peer-Prediction Makes Things Worse
摘要：
在许多情况下，评估感兴趣的对象的有效方法是从分散的个体收集评价并将这些评价聚合在一起。一些例子是分类在线内容和通过同行评分评估学生作业。对于这个数据科学问题，一个挑战是激励参与者仔细进行此类评估并诚实地报告，特别是在这样做成本高昂的情况下。现有的方法，特别是同行预测机制，可以激励在平衡中说实话。然而，它们也会产生一种均衡，在这种均衡中，代理人不支付准确评估所需的成本，因此无法获得有用的信息。我们表明，这个问题是不可避免的，每当代理人能够协调使用低成本的信号有关的项目正在评估（例如，文本标签或图片）。然后，我们考虑规避这个问题的方法，通过比较代理人的报告，真值，这是在实践中，当存在可信的评估-如助教在同行评分的情况下-谁可以执行数量有限的公正（但嘈杂）的评价。当然，当这样的真值可用时，一种更简单的方法也是可能的：基于与具有一定概率的真值的一致性来奖励每个代理，否则无条件地奖励代理。令人惊讶的是，我们发现，更简单的机制实现了更强的激励保证给予更少的访问真值比一个大的同行预测机制。
英文摘要：
In many settings, an effective way of evaluating objects of interest is to collect evaluations from dispersed individuals and to aggregate these evaluations together. Some examples are categorizing online content and evaluating student assignments via peer grading. For this data science problem, one challenge is to motivate participants to conduct such evaluations carefully and to report them honestly, particularly when doing so is costly. Existing approaches, notably peer-prediction mechanisms, can incentivize truth telling in equilibrium. However, they also give rise to equilibria in which agents do not pay the costs required to evaluate accurately, and hence fail to elicit useful information. We show that this problem is unavoidable whenever agents are able to coordinate using low-cost signals about the items being evaluated (e.g., text labels or pictures). We then consider ways of circumventing this problem by comparing agents' reports to ground truth, which is available in practice when there exist trusted evaluators---such as teaching assistants in the peer grading scenario---who can perform a limited number of unbiased (but noisy) evaluations. Of course, when such ground truth is available, a simpler approach is also possible: rewarding each agent based on agreement with ground truth with some probability, and unconditionally rewarding the agent otherwise. Surprisingly, we show that the simpler mechanism achieves stronger incentive guarantees given less access to ground truth than a large set of peer-prediction mechanisms.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
