arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2101.07577cs.LGcs.IR
收藏

推荐系统的可学习嵌入尺寸

Learnable Embedding Sizes for Recommender Systems

University of Electronic Science and Technology of China(电子科技大学);University College London(伦敦大学学院);Tsinghua University(清华大学)

Siyi Liu, Chen Gao, Yihong Chen, Depeng Jin, Yong Li

摘要:

基于嵌入的表示学习通常用于深度学习推荐模型中,以将原始稀疏特征映射到密集向量。传统的嵌入方式,分配一个统一的大小,所有的功能有两个问题。首先,众多的功能不可避免地导致巨大的嵌入表,从而导致高内存使用成本。其次,它可能会导致过度拟合的问题,这些功能,不需要太大的表示能力。现有的工作,试图解决这个问题,总是导致推荐性能显着下降,或遭受难以负担的训练时间成本的限制。在本文中,我们提出了一种新的方法,命名为PEP(插件嵌入修剪的缩写),以减少嵌入表的大小,同时避免了推荐准确率的下降。PEP修剪嵌入参数,其中修剪阈值可以从数据中自适应地学习。因此,我们可以自动获得一个混合维嵌入方案,通过修剪冗余参数的每个功能。PEP是一个通用的框架,可以插件各种基本的推荐模型。大量实验表明,该算法可以有效地减少嵌入参数,提高基模型的性能。具体来说,它实现了强大的推荐性能,同时减少了97-99%的参数。至于计算成本,PEP只带来了额外的20-30%的时间成本相比,基本模型。代码可在https://github.com/ssui-liu/learnable-embed-sizes-for-RecSys上获得。

英文摘要:

The embedding-based representation learning is commonly used in deep learning recommendation models to map the raw sparse features to dense vectors. The traditional embedding manner that assigns a uniform size to all features has two issues. First, the numerous features inevitably lead to a gigantic embedding table that causes a high memory usage cost. Second, it is likely to cause the over-fitting problem for those features that do not require too large representation capacity. Existing works that try to address the problem always cause a significant drop in recommendation performance or suffers from the limitation of unaffordable training time cost. In this paper, we proposed a novel approach, named PEP (short for Plug-in Embedding Pruning), to reduce the size of the embedding table while avoiding the drop of recommendation accuracy. PEP prunes embedding parameter where the pruning threshold(s) can be adaptively learned from data. Therefore we can automatically obtain a mixed-dimension embedding-scheme by pruning redundant parameters for each feature. PEP is a general framework that can plug in various base recommendation models. Extensive experiments demonstrate it can efficiently cut down embedding parameters and boost the base model's performance. Specifically, it achieves strong recommendation performance while reducing 97-99% parameters. As for the computation cost, PEP only brings an additional 20-30% time cost compared with base models. Codes are available at https://github.com/ssui-liu/learnable-embed-sizes-for-RecSys.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑