Saved in:
Bibliographic Details
Main Authors: Wang, Boxin, Zhang, Yibo Jacky, Cao, Yuan, Li, Bo, McMahan, H. Brendan, Oh, Sewoong, Xu, Zheng, Zaheer, Manzil
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2305.12132
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910408391000064
author Wang, Boxin
Zhang, Yibo Jacky
Cao, Yuan
Li, Bo
McMahan, H. Brendan
Oh, Sewoong
Xu, Zheng
Zaheer, Manzil
author_facet Wang, Boxin
Zhang, Yibo Jacky
Cao, Yuan
Li, Bo
McMahan, H. Brendan
Oh, Sewoong
Xu, Zheng
Zaheer, Manzil
contents We study (differentially) private federated learning (FL) of language models. The language models in cross-device FL are relatively small, which can be trained with meaningful formal user-level differential privacy (DP) guarantees when massive parallelism in training is enabled by the participation of a moderate size of users. Recently, public data has been used to improve privacy-utility trade-offs for both large and small language models. In this work, we provide a systematic study of using large-scale public data and LLMs to help differentially private training of on-device FL models, and further improve the privacy-utility tradeoff by techniques of distillation. Moreover, we propose a novel distribution matching algorithm with theoretical grounding to sample public data close to private data distribution, which significantly improves the sample efficiency of (pre-)training on public data. The proposed method is efficient and effective for training private models by taking advantage of public data, especially for customized on-device architectures that do not have ready-to-use pre-trained models.
format Preprint
id arxiv_https___arxiv_org_abs_2305_12132
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Can Public Large Language Models Help Private Cross-device Federated Learning?
Wang, Boxin
Zhang, Yibo Jacky
Cao, Yuan
Li, Bo
McMahan, H. Brendan
Oh, Sewoong
Xu, Zheng
Zaheer, Manzil
Machine Learning
We study (differentially) private federated learning (FL) of language models. The language models in cross-device FL are relatively small, which can be trained with meaningful formal user-level differential privacy (DP) guarantees when massive parallelism in training is enabled by the participation of a moderate size of users. Recently, public data has been used to improve privacy-utility trade-offs for both large and small language models. In this work, we provide a systematic study of using large-scale public data and LLMs to help differentially private training of on-device FL models, and further improve the privacy-utility tradeoff by techniques of distillation. Moreover, we propose a novel distribution matching algorithm with theoretical grounding to sample public data close to private data distribution, which significantly improves the sample efficiency of (pre-)training on public data. The proposed method is efficient and effective for training private models by taking advantage of public data, especially for customized on-device architectures that do not have ready-to-use pre-trained models.
title Can Public Large Language Models Help Private Cross-device Federated Learning?
topic Machine Learning
url https://arxiv.org/abs/2305.12132