Experimental Analysis of Large-scale Learnable Vector Storage Compression

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Hailin, Zhao, Penghao, Miao, Xupeng, Shao, Yingxia, Liu, Zirui, Yang, Tong, Cui, Bin
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929241584566272
author Zhang, Hailin
Zhao, Penghao
Miao, Xupeng
Shao, Yingxia
Liu, Zirui
Yang, Tong
Cui, Bin
author_facet Zhang, Hailin
Zhao, Penghao
Miao, Xupeng
Shao, Yingxia
Liu, Zirui
Yang, Tong
Cui, Bin
contents Learnable embedding vector is one of the most important applications in machine learning, and is widely used in various database-related domains. However, the high dimensionality of sparse data in recommendation tasks and the huge volume of corpus in retrieval-related tasks lead to a large memory consumption of the embedding table, which poses a great challenge to the training and deployment of models. Recent research has proposed various methods to compress the embeddings at the cost of a slight decrease in model quality or the introduction of other overheads. Nevertheless, the relative performance of these methods remains unclear. Existing experimental comparisons only cover a subset of these methods and focus on limited metrics. In this paper, we perform a comprehensive comparative analysis and experimental evaluation of embedding compression. We introduce a new taxonomy that categorizes these techniques based on their characteristics and methodologies, and further develop a modular benchmarking framework that integrates 14 representative methods. Under a uniform test environment, our benchmark fairly evaluates each approach, presents their strengths and weaknesses under different memory budgets, and recommends the best method based on the use case. In addition to providing useful guidelines, our study also uncovers the limitations of current methods and suggests potential directions for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2311_15578
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Experimental Analysis of Large-scale Learnable Vector Storage Compression
Zhang, Hailin
Zhao, Penghao
Miao, Xupeng
Shao, Yingxia
Liu, Zirui
Yang, Tong
Cui, Bin
Machine Learning
Databases
Information Retrieval
Learnable embedding vector is one of the most important applications in machine learning, and is widely used in various database-related domains. However, the high dimensionality of sparse data in recommendation tasks and the huge volume of corpus in retrieval-related tasks lead to a large memory consumption of the embedding table, which poses a great challenge to the training and deployment of models. Recent research has proposed various methods to compress the embeddings at the cost of a slight decrease in model quality or the introduction of other overheads. Nevertheless, the relative performance of these methods remains unclear. Existing experimental comparisons only cover a subset of these methods and focus on limited metrics. In this paper, we perform a comprehensive comparative analysis and experimental evaluation of embedding compression. We introduce a new taxonomy that categorizes these techniques based on their characteristics and methodologies, and further develop a modular benchmarking framework that integrates 14 representative methods. Under a uniform test environment, our benchmark fairly evaluates each approach, presents their strengths and weaknesses under different memory budgets, and recommends the best method based on the use case. In addition to providing useful guidelines, our study also uncovers the limitations of current methods and suggests potential directions for future research.
title Experimental Analysis of Large-scale Learnable Vector Storage Compression
topic Machine Learning
Databases
Information Retrieval
url https://arxiv.org/abs/2311.15578