Skip to main content
QUICK REVIEW

[论文解读] Revenue, Relevance, Arbitrage and More: Joint Optimization Framework for Search Experiences in Two-Sided Marketplaces

Andrew Stanton, Akhila Ananthram|arXiv (Cornell University)|May 15, 2019
Consumer Market Behavior and Pricing参考文献 31被引用 5
一句话总结

本文提出了一种使用进化策略的联合优化框架,用于在Etsy等双边市场中同时平衡收入、相关性、公平性(基尼指数)和卖家激励指标。通过将市场级指标建模为约束条件,并端到端训练策略网络,该方法在无需级联手动调优的情况下,实现了对多样化业务目标的优化,在公平性和相关性方面取得了统计上显著的提升,同时保持了较高的NDCG和GMV性能。

ABSTRACT

Two-sided marketplaces such as eBay, Etsy and Taobao have two distinct groups of customers: buyers who use the platform to seek the most relevant and interesting item to purchase and sellers who view the same platform as a tool to reach out to their audience and grow their business. Additionally, platforms have their own objectives ranging from growing both buyer and seller user bases to revenue maximization. It is not difficult to see that it would be challenging to obtain a globally favorable outcome for all parties. Taking the search experience as an example, any interventions are likely to impact either buyers or sellers unfairly to course correct for a greater perceived need. In this paper, we address how a company-aligned search experience can be provided with competing business metrics that E-commerce companies typically tackle. As far as we know, this is a pioneering work to consider multiple different aspects of business indicators in two-sided marketplaces to optimize a search experience. We demonstrate that many problems are difficult or impossible to decompose down to credit assigned scores on individual documents, rendering traditional methods inadequate. Instead, we express market-level metrics as constraints and discuss to what degree multiple potentially conflicting metrics can be tuned to business needs. We further explore the use of policy learners in the form of Evolutionary Strategies to jointly optimize both group-level and market-level metrics simultaneously, side-stepping traditional cascading methods and manual interventions. We empirically evaluate the effectiveness of the proposed method on Etsy data and demonstrate its potential with insights.

研究动机与目标

  • 解决双边市场中买家相关性、卖家激励与平台公平性之间常发生冲突的搜索体验优化挑战。
  • 克服传统排序学习方法无法分解复杂市场级指标(如基尼指数或激励公平性)的局限性。
  • 开发一个统一框架,联合优化多个相互冲突的业务指标,而无需依赖级联启发式方法或人工干预。
  • 证明通过进化策略进行策略学习可有效调整搜索策略,以满足多样且相互竞争的业务需求。
  • 在真实Etsy数据上验证该方法,显示在公平性和相关性方面有可衡量的改进,同时保持核心排序性能。

提出的方法

  • 该框架将市场级指标(如基尼指数、激励公平性、收入)作为多目标优化问题中的约束条件进行建模。
  • 采用进化策略(ES)训练策略网络,直接优化多个目标,包括相关性(NDCG)、公平性(基尼)和激励一致性。
  • 策略网络使用随机价值函数(SVFs)和贪婪价值函数(GVFs)进行训练,其中前者表现出更好的泛化能力和更低的方差。
  • 该方法使用重要性加权来平衡各项目标之间的权衡,实现市场级指标的平滑且渐进的调优。
  • 采用点对点与群体级优化相结合的混合方法,实现单个文档相关性与整体市场健康状况的联合提升。
  • 该框架通过学习单一策略来同时满足多样且可能冲突的业务目标,避免了传统级联方法的使用。

实验结果

研究问题

  • RQ1是否可以在不依赖点对点损失函数的情况下,有效优化搜索排序中的市场级指标(如公平性(基尼指数)和卖家激励对齐)?
  • RQ2进化策略在双边市场中联合优化相关性、收入和公平性等相互冲突的目标方面,其有效程度如何?
  • RQ3在优化市场级指标时,随机价值函数与静态价值函数在泛化能力和稳定性方面有何差异?
  • RQ4所提出的框架在多大程度上可以减少搜索排序系统中对人工调优和级联干预的依赖?
  • RQ5不同指标的重要性加权如何影响搜索结果中公平性与相关性之间的权衡?

主要发现

  • SG-ES-{0.17,0.17}变体在测试集上相较于MMR-LambdaMART基线,NDCG提升了1.5%(0.652 vs. 0.642),基尼指数提升了1.4%(0.911 vs. 0.889)。
  • 随机策略变体(SG-ES)相比贪婪变体将测试方差降低了50%,表现出更好的泛化能力和稳定性。
  • 在SG-ES-{0.17,0.17}配置下,激励公平性提升了14.5%(0.543 vs. 0.466),表明与卖家激励的对齐性更强。
  • 重要性加权实现了市场级指标的平滑渐进调优,基尼指数从基线的0.795提升至最优加权下的0.911。
  • SG-ES变体在相关性(NDCG)和公平性(基尼)方面均优于所有基线,测试NDCG较次佳基线提升1.2%,基尼指数也提升1.2%。
  • 该方法表明,通过策略学习可可靠地影响市场级指标(如基尼指数和激励公平性),从而减少对人工或启发式干预的依赖。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。