Deep Learning Recommendation Model (DLRM)

Optimizing Embedding Tables

  • UGache: A Unified GPU Cache for Embedding-based Deep Learning (SOSP 2023) [Personal Notes] [Paper]

    • SJTU

    • A unified multi-GPU cache system.

    • Used for GNN training and DLR inference.

  • Bagpipe: Accelerating Deep Recommendation Model Training (SOSP 2023) [Paper]

    • UW-Madison & UChicago

  • AdaEmbed: Adaptive Embedding for Large-Scale Recommendation Models (OSDI 2023) [Paper]

    • UMich SymbioticLab & Meta

    • In-training pruning

  • EVStore: Storage and Caching Capabilities for Scaling Embedding Tables in Deep Recommendation Systems (ASPLOS 2023) [Personal Notes] [Paper] [Code]

    • UChicago & Beijing University of Technology & Bandung Institute of Technology, Indonesia & Seagate Technology & Emory

    • A caching layer optimized for embedding access patterns.

  • DisaggRec: Architecting Disaggregated Systems for Large-Scale Personalized Recommendation (arXiv 2212.00939) [Personal Notes] [Paper]

    • Meta AI & WashU & UPenn & Cornell & Intel

    • Disaggregated system; decouple CPUs and memory resources; partition embedding tables.

Faster Model Update

  • Ekko: A Large-Scale Deep Learning Recommender System with Low-Latency Model Update (OSDI 2022) [Paper]

    • Tencent & Edinburgh

    • P2P model update dissemination.

DLRM

  • Efficient Long Sequential User Data Modeling for Click-Through Rate Prediction (DLP-KDD 2022) [Paper]

    • Alibaba

    • ETA: Efficient target attention network

    • Locality-sensitive hashing

    • Deployed on Taobao.

  • Wide & Deep Learning for Recommender Systems (DLRS 2016) [Personal Notes] [Paper]

    • Google

    • WDL: Wide & Deep model

Acronyms

  • DLRM: Deep Learning Recommendation Model

Last updated