现代化家庭信息流预排名阶段
[
[
Bella Huang | Machine Learning Engineer, Homefeed Candidate GenerationDafang He | Machine Learning Engineer, Homefeed RelevanceYuying Chen | Machine Learning Engineer, (formerly) Homefeed Candidate GenerationJames Li | Engineering Manager, Homefeed Candidate Generation
Bella Huang | 机器学习工程师，Homefeed候选生成Dafang He | 机器学习工程师，Homefeed相关性Yuying Chen | 机器学习工程师，（前）Homefeed候选生成James Li | 工程经理，Homefeed候选生成
Dylan Wang | Director, Homefeed Relevance
Dylan Wang | 主管，Homefeed相关性
Modern recommendation systems typically follow a multi-stage design involving retrieval, pre-ranking, ranking, and reranking. Pinterest home feed recommendation system has similarly adopted these strategies over the years. We are excited to announce the big milestone in this journey with a sophisticated pre-ranking layer (aka Lightweight Scoring) that significantly improved our business metrics.
现代推荐系统通常遵循多阶段设计，包括检索、预排名、排名和重新排名。Pinterest主页推荐系统多年来也采用了这些策略。我们很高兴地宣布，在这段旅程中取得了一个重要里程碑，推出了一个复杂的预排名层（即轻量级评分），显著改善了我们的业务指标。
In this blog, we’d like to share the most foundational improvements in this stage, including both end-to-end system design and model design specifically tailored for this stage. Figure 1 illustrates the end-to-end Pinterest funnel design:
在这篇博客中，我们想分享这一阶段最 基础 的改进，包括专门为这一阶段量身定制的 端到端系统设计 和 模型设计。图1展示了端到端的Pinterest漏斗设计：
Figure 1: End-to-End home feed Funnel with the focus of this work
图1：端到端主页推荐漏斗，本文的重点
Limitations of the Initial Design
初始设计的局限性
Many industrial level pre-ranking layers adopt a two-tower based approach [1][2]. In [1], we described how Pinterest home feed built our first generation of light-weight rankers. One key part is that in this design, the light-ranker runs separately on each retrieval source output. We then aggregate all the results from each retrieval source in the home feed lo...