[论文解读] Beyond Match Maximization and Fairness: Retention-Optimized Two-Sided Matching

Ren Kishimoto, Rikiya Takehi+5 more|arXiv (Cornell University)|Feb 17, 2026

Evolutionary Psychology and Human Behavior被引用 0

一句话总结

论文定义了保留率优化的双边匹配并引入 MRet，一种动态学习排序算法，在平台两端最大化用户留存，在合成数据和真实约会数据上得到验证。

ABSTRACT

On two-sided matching platforms such as online dating and recruiting, recommendation algorithms often aim to maximize the total number of matches. However, this objective creates an imbalance, where some users receive far too many matches while many others receive very few and eventually abandon the platform. Retaining users is crucial for many platforms, such as those that depend heavily on subscriptions. Some may use fairness objectives to solve the problem of match maximization. However, fairness in itself is not the ultimate objective for many platforms, as users do not suddenly reward the platform simply because exposure is equalized. In practice, where user retention is often the ultimate goal, casually relying on fairness will leave the optimization of retention up to luck. In this work, instead of maximizing matches or axiomatically defining fairness, we formally define the new problem setting of maximizing user retention in two-sided matching platforms. To this end, we introduce a dynamic learning-to-rank (LTR) algorithm called Matching for Retention (MRet). Unlike conventional algorithms for two-sided matching, our approach models user retention by learning personalized retention curves from each user's profile and interaction history. Based on these curves, MRet dynamically adapts recommendations by jointly considering the retention gains of both the user receiving recommendations and those who are being recommended, so that limited matching opportunities can be allocated where they most improve overall retention. Naturally but importantly, empirical evaluations on synthetic and real-world datasets from a major online dating platform show that MRet achieves higher user retention, since conventional methods optimize matches or fairness rather than retention.

研究动机与目标

• 解决在双边平台中最大化匹配导致失衡和流失的问题。
• 提出将留存作为主要目标，而非公平性或匹配数量。
• 从用户画像与历史中学习个性化留存曲线。
• 开发 MRet，使稀缺的匹配机会分配以最大化整体留存。
• 在合成数据和真实世界的约会数据上展示更优的留存表现。

提出的方法

• 为每个用户端正式定义保留函数 f(x,m)，并将留存概率建模为累计匹配数 m 的函数。
• 推导出一个最大化留存的排序目标（式 (Eq. 9)），该目标对接收方的留存增益与推荐方的留存增益求和。
• 引入 MRet 排序器，通过基于凹性下界的近似和 Score(y) 函数（式 (Eq. 13)）使 NP-hard 目标实现 O(N log N) 的排序。
• 利用 Jensen 型界限（引理 1 和 2）将目标分解并界定，使优化可行。
• 训练回归模型从上下文（XGBoost）估计 f，并利用观察到的交互来学习留存动力学。
• 用合成数据和真实的日本在线约会数据集进行验证，代码在补充材料中提供。

实验结果

研究问题

• RQ1 留存而非匹配或公平性，能否在双边匹配中直接优化？
• RQ2 动态学习排序方法（MRet）是否相较于 Max Match 或 FairCo 等基线在长期留存上有所提升？
• RQ3 NP-hard 的留存优化是否可以高效近似且不牺牲性能？
• RQ4 在不同条件下，MRet 在合成数据和真实世界约会平台数据上的表现如何？
• RQ5 当留存函数的凹性假设偏离时，MRet 是否仍具鲁棒性？

主要发现

• MRet 的用户留存高于 Max Match、Uniform 和 FairCo 等基线，同时使用的匹配量约为 Max Match 的 70%。
• 在合成实验中，MRet 在不同受欢迎程度偏斜（kappa）条件下的留存优于基线。
• 在真实世界数据中，在极度稀疏的匹配场景下，MRet 仍实现最高留存，而公平性方法则失效。
• FairCo 可能因曝光公平性与实际留存需求之间的错位而在留存优化上表现不佳。
• 即使在真实数据中违背凹性假设，MRet 仍然有效，显示出鲁棒性。