arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.02905cs.GT
收藏

评分规则的优化

Optimization of Scoring Rules

Yale University(耶鲁大学);Northwestern University(西北大学);Toyota Technological Institute at Chicago(芝加哥丰田技术研究所)

Jason D. Hartline, Yingkai Li, Liren Shan, Yifan Wu

AI总结:

该研究刻画了激励智能体获取并如实上报信息的最优评分规则,明确了单/对称多维度问题下的规则形式,验证其对非对称分布近似最优,指出传统标准评分规则效果欠佳。

摘要:

我们刻画了能激励智能体获取信息并向委托方如实上报信息的最优奖励函数(即评分规则)。在单维度问题中,最优评分规则允许智能体进行简单的二元押注;在对称多维度问题中,智能体可选择信号最出人意料的维度作为评分依据。这种评分规则形式对于非对称分布仍能保持近似最优。这些结果表明,在多维度问题中,将激励机制与获取高质量信息挂钩至关重要。相比之下,二次评分规则等标准评分规则,或是单维度评分规则的平均值,其效果可能远非最优。

英文摘要:

We characterize the optimal reward functions (scoring rules) that incentivize an agent to acquire information and report it truthfully to the principal. The optimal scoring rules let the agent make a simple binary bet in single-dimensional problems, and choose the dimension with the most surprising signal to be scored on in symmetric multi-dimensional problems. This scoring rule format remains approximately optimal for asymmetric distributions. These results demonstrate the importance of linking incentives to obtain high-quality information in multi-dimensional problems. In contrast, standard scoring rules like the quadratic scoring rule, or averages of single-dimensional scoring rules can be far from optimal.

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。

↑