arXiv：cs.LG· Ermis Soumalias, Richard Mudd, Abbas Zaidi·· 4 小时前AI 评分36
在线实验中的激励对齐：样本拆分与收缩机制如何解决平台实验的委托代理冲突
Incentive Alignment in Online Experimentation
AI 导读
针对在线平台实验中实验者自主决定测试假设、且按平均处理效应获得奖励所导致的激励错位问题，研究将实验设计重新表述为激励设计问题，提出样本拆分和收缩两种机制。样本拆分以有限流量成本实现激励完全对齐，收缩机制不消耗额外流量且能保证负期望效应的干预严格无利可图。该委托代理冲突无法通过显著性阈值和流量预算等集中式手段解决。
正文
Abstract:Evaluating the causal effect of new features is a central goal for online platforms. While recent literature addresses limited testing traffic via centralized portfolio optimization, this perspective abstracts away a critical institutional reality: experimentation is operationally decentralized. The experimenters who develop new features also dictate which hypotheses to test, and they are typically rewarded based on empirical average treatment effects that are prone to upward bias. Left unchecked, this principal-agent conflict can severely erode platform value, a structural failure that conventional centralized levers, such as significance thresholds and traffic budgets, cannot resolve. By reframing experimentation as an incentive design problem, we demonstrate that two practical mechanisms, sample splitting and shrinkage, can effectively bridge this gap. Sample splitting aligns incentives perfectly at a bounded traffic cost, while shrinkage consumes no additional traffic and guarantees that interventions with negative expected effects are strictly unprofitable to field.
| Subjects: | Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.05922 [cs.GT] |
| (or arXiv:2610.05922v2 [cs.GT] for this version) | |
| https://doi.org/10.48550/arXiv.2610.05922 arXiv-issued DOI via DataCite |
Submission history
From: Ermis Soumalias Dr. [view email]
[v1]
Mon, 5 Oct 2026 07:41:43 UTC (60 KB)
[v2]
Tue, 6 Oct 2026 08:44:16 UTC (60 KB)
来源：arXiv：cs.LG · arxiv.org