Outsourcing Training without Uploading Data via Efficient Collaborative Open-Source Sampling
Not on the map. Why?
Abstract (arXiv)
As deep learning blooms with growing demand for computation and data resources, outsourcing model training to a powerful cloud server becomes an attractive alternative to training at a low-power and cost-effective end device. Traditional outsourcing requires uploading device data to the cloud server, which can be infeasible in many real-world applications due to the often sensitive nature of the collected data and the limited communication bandwidth. To tackle these challenges, we propose to leverage widely available open-source data, which is a massive dataset collected from public and heterogeneous sources (e.g., Internet images). We develop a novel strategy called Efficient Collaborative Open-source Sampling (ECOS) to construct a proximal proxy dataset from open-source data for cloud training, in lieu of client data. ECOS probes open-source data on the cloud server to sense the distribution of client data via a communication- and computation-efficient sampling process, which only communicates a few compressed public features and client scalar responses. Extensive empirical studies show that the proposed ECOS improves the quality of automated client labeling, model compression, and label outsourcing when applied in various learning scenarios.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 28 of its 60 references that are in the map. First posted on arXiv 23 Oct 2022.
Builds on
28 of 60 references.
References · 28
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2015Distilling the Knowledge in a Neural Networkcited by 10,324
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2014Deep Learning Face Attributes in the Wildcited by 3,693
- 2016Pointer Sentinel Mixture Modelscited by 3,096
- 2021Convex optimizationcited by 2,515
- 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
- 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
- 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,331
- 2017Renyi Differential Privacycited by 650
- 2021FedBN: Federated Learning on Non-IID Features via Local Batch Normalizationcited by 571
- 2018Split learning for health: Distributed deep learning without sharing raw patient datacited by 395
16 more
- 2016Semi-supervised Knowledge Transfer for Deep Learning from Private Training Datacited by 348
- 2018Scalable Private Learning with PATEcited by 269
- 2018Subsampled R'enyi Differential Privacy and Analytical Moments Accountantcited by 236
- 2020Privacy and Robustness in Federated Learning: Attacks and Defensescited by 166
- 2019Federated Adversarial Domain Adaptationcited by 140
- 2018In-Edge AI: Intelligentizing Mobile Edge Computing, Caching and Communication by Federated Learningcited by 80
- 2020What Can Be Transferred: Unsupervised Domain Adaptation for Endoscopic Lesions Segmentationcited by 50
- 2019Limits of Private Learning with Access to Public Datacited by 39
- 2021Leveraging Public Data for Practical Private Query Releasecited by 39
- 2021Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AIcited by 39
- 2022Efficient Split-Mix Federated Learning for On-Demand and In-Situ Customizationcited by 35
- 2018Deep Private-Feature Extractioncited by 30
- 2021Improving Contrastive Learning on Imbalanced Seed Data via Open-World Samplingcited by 30
- 2019Positive-Unlabeled Compression on the Cloudcited by 27
- 2020Neural Data Server: A Large-Scale Search Engine for Transfer Learning Datacited by 27
- 2022CalFAT: Calibrated Federated Adversarial Training with Label Skewnesscited by 20
Built on by
Citing papers per year: 2023: 1; 2024: 1; 2025: 1.
3 papers.