@inproceedings{48d7b2f0564c42029d91f363d63c77f0,
title = "Alternate Model Growth and Pruning for Efficient Training of Recommendation Systems",
abstract = "Deep learning recommendation systems at scale have provided remarkable gains through increasing model capacity (i.e. wider and deeper neural networks), but it comes at significant training cost and infrastructure cost. Model pruning is an effective technique to reduce computation overhead for deep neural networks by removing redundant parameters. However, modern recommendation systems are still thirsty for model capacity due to the demand for handling big data. Thus, pruning a recommendation model at scale results in a smaller model capacity and consequently lower accuracy. To reduce computation cost without sacrificing model capacity, we propose a dynamic training scheme, namely alternate model growth and pruning, to alternatively construct and prune weights in the course of training. Our method leverages structured sparsification to reduce computational cost without hurting the model capacity at the end of offline training so that a full-size model is available in the recurring training stage to learn new data in real time. To the best of our knowledge, this is the first work to provide in-depth experiments and discussion of applying structural dynamics to recommendation systems at scale to reduce training cost. The proposed method is validated with an open-source deep learning recommendation model (DLRM) and state-of-the-art industrial-scale production models.",
keywords = "Deep learning, Deep neural network, Efficient model training, Model growth, Model pruning, Recommendation system",
author = "Xiaocong Du and Bhargav Bhushanam and Jiecao Yu and Dhruv Choudhary and Tianxiang Gao and Sherman Wong and Louis Feng and Jongsoo Park and Yu Cao and Arun Kejariwal",
note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021 ; Conference date: 13-12-2021 Through 16-12-2021",
year = "2021",
doi = "10.1109/ICMLA52953.2021.00229",
language = "English (US)",
series = "Proceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1421--1428",
editor = "Wani, {M. Arif} and Sethi, {Ishwar K.} and Weisong Shi and Guangzhi Qu and Raicu, {Daniela Stan} and Ruoming Jin",
booktitle = "Proceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021",
}