功能强大的 A/B 测试指标及查找方法

Olivier Jeunen, Shubham Baweja, Neeti Pokharna, Aleksei Ustimenko
{"title":"功能强大的 A/B 测试指标及查找方法","authors":"Olivier Jeunen, Shubham Baweja, Neeti Pokharna, Aleksei Ustimenko","doi":"arxiv-2407.20665","DOIUrl":null,"url":null,"abstract":"Online controlled experiments, colloquially known as A/B-tests, are the bread\nand butter of real-world recommender system evaluation. Typically, end-users\nare randomly assigned some system variant, and a plethora of metrics are then\ntracked, collected, and aggregated throughout the experiment. A North Star\nmetric (e.g. long-term growth or revenue) is used to assess which system\nvariant should be deemed superior. As a result, most collected metrics are\nsupporting in nature, and serve to either (i) provide an understanding of how\nthe experiment impacts user experience, or (ii) allow for confident\ndecision-making when the North Star metric moves insignificantly (i.e. a false\nnegative or type-II error). The latter is not straightforward: suppose a\ntreatment variant leads to fewer but longer sessions, with more views but fewer\nengagements; should this be considered a positive or negative outcome? The question then becomes: how do we assess a supporting metric's utility\nwhen it comes to decision-making using A/B-testing? Online platforms typically\nrun dozens of experiments at any given time. This provides a wealth of\ninformation about interventions and treatment effects that can be used to\nevaluate metrics' utility for online evaluation. We propose to collect this\ninformation and leverage it to quantify type-I, type-II, and type-III errors\nfor the metrics of interest, alongside a distribution of measurements of their\nstatistical power (e.g. $z$-scores and $p$-values). We present results and\ninsights from building this pipeline at scale for two large-scale short-video\nplatforms: ShareChat and Moj; leveraging hundreds of past experiments to find\nonline metrics with high statistical power.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Powerful A/B-Testing Metrics and Where to Find Them\",\"authors\":\"Olivier Jeunen, Shubham Baweja, Neeti Pokharna, Aleksei Ustimenko\",\"doi\":\"arxiv-2407.20665\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Online controlled experiments, colloquially known as A/B-tests, are the bread\\nand butter of real-world recommender system evaluation. Typically, end-users\\nare randomly assigned some system variant, and a plethora of metrics are then\\ntracked, collected, and aggregated throughout the experiment. A North Star\\nmetric (e.g. long-term growth or revenue) is used to assess which system\\nvariant should be deemed superior. As a result, most collected metrics are\\nsupporting in nature, and serve to either (i) provide an understanding of how\\nthe experiment impacts user experience, or (ii) allow for confident\\ndecision-making when the North Star metric moves insignificantly (i.e. a false\\nnegative or type-II error). The latter is not straightforward: suppose a\\ntreatment variant leads to fewer but longer sessions, with more views but fewer\\nengagements; should this be considered a positive or negative outcome? The question then becomes: how do we assess a supporting metric's utility\\nwhen it comes to decision-making using A/B-testing? Online platforms typically\\nrun dozens of experiments at any given time. This provides a wealth of\\ninformation about interventions and treatment effects that can be used to\\nevaluate metrics' utility for online evaluation. We propose to collect this\\ninformation and leverage it to quantify type-I, type-II, and type-III errors\\nfor the metrics of interest, alongside a distribution of measurements of their\\nstatistical power (e.g. $z$-scores and $p$-values). We present results and\\ninsights from building this pipeline at scale for two large-scale short-video\\nplatforms: ShareChat and Moj; leveraging hundreds of past experiments to find\\nonline metrics with high statistical power.\",\"PeriodicalId\":501172,\"journal\":{\"name\":\"arXiv - STAT - Applications\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - STAT - Applications\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.20665\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - STAT - Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.20665","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在线控制实验,俗称 A/B 测试,是现实世界中推荐系统评估的主要手段。通常,终端用户会被随机分配到某个系统变体,然后在整个实验过程中跟踪、收集和汇总大量指标。北指标(如长期增长或收入)用于评估哪个系统变体更优。因此,收集到的大多数指标在本质上都是辅助性的,其作用是:(i) 提供对实验如何影响用户体验的理解,或 (ii) 当 "北极星 "指标移动不明显时(即出现假阴性或 II 型错误),可以做出有把握的决策。后者并不简单:假设治疗变体会导致会话次数减少但时间延长,浏览次数增加但登录次数减少;这应该被视为积极结果还是消极结果?问题就变成了:在使用 A/B 测试进行决策时,我们该如何评估辅助指标的效用?在线平台通常会在任何时间运行数十个实验。这提供了大量有关干预和处理效果的信息,可用于评估在线评估指标的效用。我们建议收集这些信息,并将其用于量化相关指标的 I 类、II 类和 III 类误差,以及统计能力的测量分布(如 z 值和 p 值)。我们介绍了为两个大型短视频平台大规模构建该管道的结果和启示:ShareChat 和 Moj;利用过去的数百次实验,找到具有高统计能力的在线指标。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Powerful A/B-Testing Metrics and Where to Find Them
Online controlled experiments, colloquially known as A/B-tests, are the bread and butter of real-world recommender system evaluation. Typically, end-users are randomly assigned some system variant, and a plethora of metrics are then tracked, collected, and aggregated throughout the experiment. A North Star metric (e.g. long-term growth or revenue) is used to assess which system variant should be deemed superior. As a result, most collected metrics are supporting in nature, and serve to either (i) provide an understanding of how the experiment impacts user experience, or (ii) allow for confident decision-making when the North Star metric moves insignificantly (i.e. a false negative or type-II error). The latter is not straightforward: suppose a treatment variant leads to fewer but longer sessions, with more views but fewer engagements; should this be considered a positive or negative outcome? The question then becomes: how do we assess a supporting metric's utility when it comes to decision-making using A/B-testing? Online platforms typically run dozens of experiments at any given time. This provides a wealth of information about interventions and treatment effects that can be used to evaluate metrics' utility for online evaluation. We propose to collect this information and leverage it to quantify type-I, type-II, and type-III errors for the metrics of interest, alongside a distribution of measurements of their statistical power (e.g. $z$-scores and $p$-values). We present results and insights from building this pipeline at scale for two large-scale short-video platforms: ShareChat and Moj; leveraging hundreds of past experiments to find online metrics with high statistical power.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信