一句话：提出混合嵌入表示与MP-Rec软硬件协同技术，动态选择嵌入表示和硬件平台以提升推荐系统性能。
深度推荐系统需在多样尾延迟目标和输入负载下提供个性化内容，但现有模型依赖 TB 级嵌入表，固定嵌入表示带来巨大内存容量与带宽需求，并限制可用系统方案。本文挑战该假设，通过刻画多种嵌入表示，揭示嵌入表示与硬件平台的协同可同时改善算法与系统性能。作者提出混合嵌入表示，以更高内存和计算开销换取更高质量嵌入；并针对其系统性能挑战提出 MP-Rec，一种利用异构性、动态选择嵌入表示与底层硬件平台的软硬件协同设计。摘要未报告具体实验数值，评估基于嵌入表示表征分析以及混合表示和 MP-Rec 的算法与系统性能。
Deep learning recommendation systems serve personalized content under diverse tail-latency targets and input-query loads. In order to do so, state-of-the-art recommendation models rely on terabyte-scale embedding tables to learn user preferences over large bodies of contents. The reliance on a fixed embedding representation of embedding tables not only imposes significant memory capacity and bandwidth requirements but also limits the scope of compatible system solutions. This paper challenges the assumption of fixed embedding representations by showing how synergies between embedding representations and hardware platforms can lead to improvements in both algorithmic- and system performance. Based on our characterization of various embedding representations, we propose a hybrid embedding representation that achieves higher quality embeddings at the cost of increased memory and compute requirements. To address the system performance challenges of the hybrid representation, we propose MP-Rec — a co-design technique that exploits heterogeneity and dynamic selection of embedding representations and underlying hardware platforms.
@inproceedings{paperbot10086,
title = {MP-Rec: Hardware-Software Co-design to Enable Multi-path Recommendation},
author = {Samuel Hsia and Udit Gupta and Bilge Acun and Newsha Ardalani and Pan Zhong and Gu-Yeon Wei and David Brooks and Carole-Jean Wu},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3582016.3582068}
}