LiRank: Industrial Large Scale Ranking Models at LinkedIn

We present LiRank, a large-scale ranking framework at LinkedIn that brings to production state-of-the-art modeling architectures and optimization methods. We unveil several modeling improvements, including Residual DCN, which adds attention and residual connections to the famous DCNv2 architecture....

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-08
Main Authors: Borisyuk, Fedor, Zhou, Mingzhou, Song, Qingquan, Zhu, Siyu, Tiwana, Birjodh, Parameswaran, Ganesh, Dangi, Siddharth, Hertel, Lars, Xiao, Qiang, Hou, Xiaochen, Ouyang, Yunbo, Gupta, Aman, Singh, Sheallika, Liu, Dan, Cheng, Hailing, Le, Lei, Hung, Jonathan, Sathiya Keerthi, Wang, Ruoyan, Zhang, Fengyu, Kothari, Mohit, Chen, Zhu, Sun, Daqi, Dai, Yun, Luan, Xun, Zhu, Sirou, Wang, Zhiwei, Daftary, Neil, Shen, Qianqi, Jiang, Chengming, Wei, Haichao, Varshney, Maneesh, Ghoting, Amol, Ghosh, Souvik
Format: Article
Language:English
Subjects:
Modelling
Production methods
Ranking
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present LiRank, a large-scale ranking framework at LinkedIn that brings to production state-of-the-art modeling architectures and optimization methods. We unveil several modeling improvements, including Residual DCN, which adds attention and residual connections to the famous DCNv2 architecture. We share insights into combining and tuning SOTA architectures to create a unified model, including Dense Gating, Transformers and Residual DCN. We also propose novel techniques for calibration and describe how we productionalized deep learning based explore/exploit methods. To enable effective, production-grade serving of large ranking models, we detail how to train and compress models using quantization and vocabulary compression. We provide details about the deployment setup for large-scale use cases of Feed ranking, Jobs Recommendations, and Ads click-through rate (CTR) prediction. We summarize our learnings from various A/B tests by elucidating the most effective technical approaches. These ideas have contributed to relative metrics improvements across the board at LinkedIn: +0.5% member sessions in the Feed, +1.76% qualified job applications for Jobs search and recommendations, and +4.3% for Ads CTR. We hope this work can provide practical insights and solutions for practitioners interested in leveraging large-scale deep ranking systems.
ISSN:2331-8422