Federated Ensembles: a literature review

Author:

Daalen Florian1,Ippel Lianne2,Dekker Andre1,Bermejo Inigo1

Affiliation:

1. Maastricht University

2. Centraal Bureau voor de Statistiek

Abstract

Abstract Federated learning (FL) allows machine learning algorithms to be applied to decentralized data when data sharing is not an option due to privacy concerns. Ensemble-based learning works by training multiple (weak) classifiers whose output is aggregated. Federated ensembles are ensembles applied to a federated setting, where each classifier in the ensemble is trained on one data location. The aim of this review is to provide an overview of the published literature on federated ensembles, their applications, the methods used, the challenges faced, the proposed solutions and their comparative performance. We searched for publications on federated ensembles on five databases (ACM Digital Library, IEEE, arXiv, Google scholar and Scopus) published after 2016. We found 26 articles describing studies either proposing federated ensemble applications or comparing federated ensembles to other federated learning approaches. Federated ensembles were used for a wide varied applications beyond classification. Advocates of federated ensemble mentioned their ability to handle local biases in data. In comparison to federated learning approaches, federated ensembles underperformed in small sample sizes and highly class imbalanced settings. Only 10 articles discussed privacy guarantees or additional privacy preserving techniques. Federated ensembles represent an interesting alternative to federated averaging algorithms that is inherently privacy preserving. They have proved their versatility but remain underutilized.

Publisher

Research Square Platform LLC

Reference46 articles.

1. Li, Li and Fan, Yuxi and Tse, Mike and Lin, Kuo-Yi (2020) A review of applications in federated learning. Computers & Industrial Engineering 149: 106854 https://doi.org/10.1016/j.cie.2020.106854, ScienceDirect Snapshot:C\:\\Users\\p70074073\\Zotero\\storage\\487GUWX5\\S0360835220305532.html:text/html, Citation analysis, Federated learning, Literature review, Research front, November, 2021-03-03, en, Federated Learning (FL) is a collaboratively decentralized privacy-preserving technology to overcome challenges of data silos and data sensibility. Exactly what research is carrying the research momentum forward is a question of interest to research communities as well as industrial engineering. This study reviews FL and explores the main evolution path for issues exist in FL development process to advance the understanding of FL. This study aims to review prevailing application in industrial engineering to guide for the future landing application. This study also identifies six research fronts to address FL literature and help advance our understanding of FL for future optimization. This study contributes to conclude application in industrial engineering and computer science and summarize a review of applications in FL., 0360-8352

2. Cao, Xiaoyu and Jia, Jinyuan and Gong, Neil Zhenqiang (2021) Provably {Secure} {Federated} {Learning} against {Malicious} {Clients}. Proceedings of the AAAI Conference on Artificial Intelligence 35(8): 6885--6893 https://doi.org/10.1609/aaai.v35i8.16849, Full Text PDF:C\:\\Users\\p70074073\\Zotero\\storage\\YQG9FS7S\\Cao et al. - 2021 - Provably Secure Federated Learning against Malicio.pdf:application/pdf, Ensemble Methods, Number: 8, May, 2022-11-25, en, Federated learning enables clients to collaboratively learn a shared global model without sharing their local training data with a cloud server. However, malicious clients can corrupt the global model to predict incorrect labels for testing examples. Existing defenses against malicious clients leverage Byzantine-robust federated learning methods. However, these methods cannot provably guarantee that the predicted label for a testing example is not affected by malicious clients. We bridge this gap via ensemble federated learning. In particular, given any base federated learning algorithm, we use the algorithm to learn multiple global models, each of which is learnt using a randomly selected subset of clients. When predicting the label of a testing example, we take majority vote among the global models. We show that our ensemble federated learning with any base federated learning algorithm is provably secure against malicious clients. Specifically, the label predicted by our ensemble global model for a testing example is provably not affected by a bounded number of malicious clients. Moreover, we show that our derived bound is tight. We evaluate our method on MNIST and Human Activity Recognition datasets. For instance, our method can achieve a certified accuracy of 88% on MNIST when 20 out of 1,000 clients are malicious., 2374-3468, Copyright (c) 2021 Association for the Advancement of Artificial Intelligence

3. Parmar, Payal V. and Padhar, Shraddha B. and Patel, Shafika N. and Bhatt, Niyatee I. and Jhaveri, Rutvij H. (2014) Survey of Various Homomorphic Encryption algorithms and Schemes. International Journal of Computer Applications 91(8): 26--32 https://doi.org/10.5120/15902-5081, V.Parmar et al. - 2014 - Survey of Various Homomorphic Encryption algorithm.pdf:C\:\\Users\\p70074073\\Zotero\\storage\\DXTUHNEX\\V.Parmar et al. - 2014 - Survey of Various Homomorphic Encryption algorithm.pdf:application/pdf, english, 2014-04-18, April, 2021-07-02, {IJCA}, International Journal of Computer Applications, Homomorphic encryption is the encryption scheme which means the operations on the encrypted data. Homomorphic encryption can be applied in any system by using various public key algorithms. When the data is transferred to the public area, there are many encryption algorithms to secure the operations and the storage of the data. But to process data located on remote server and to preserve privacy, homomorphic encryption is useful that allows the operations on the cipher text, which can provide the same results after calculations as the working directly on the raw data. In this paper, the main focus is on public key cryptographic algorithms based on homomorphic encryption scheme for preserving security. The case study on various principles and properties of homomorphic encryption is given and then various homomorphic algorithms using asymmetric key systems such as {RSA}, {ElGamal}, Paillier algorithms as well as various homomorphic encryption schemes such as {BrakerskiGentry}-Vaikuntanathan ({BGV}), Enhanced homomorphic Cryptosystem ({EHC}), Algebra homomorphic encryption scheme based on updated {ElGamal} ({AHEE}), Non-interactive exponential homomorphic encryption scheme ({NEHE}) are investigated., 09758887

4. Ahn, S. and Özg ür, A. and Pilanci, M. (2020) Global Multiclass Classification and Dataset Construction via Heterogeneous Local Experts. {IEEE} Journal on Selected Areas in Information Theory 1(3): 870--883 https://doi.org/10.1109/JSAIT.2020.3041804, Submitted Version:C\:\\Users\\p70074073\\Zotero\\storage\\RE6ZFAQ8\\Ahn et al. - 2020 - Global Multiclass Classification and Dataset Const.pdf:application/pdf, 2020, November, {IEEE} Journal on Selected Areas in Information Theory, 2641-8770

5. Anagnostopoulos, C. (2022) Edge-centric inferential modeling & analytics. Journal of Network and Computer Applications 164 https://doi.org/10.1016/j.jnca.2020.102696, Full Text:C\:\\Users\\p70074073\\Zotero\\storage\\LA5WQZ9D\\Anagnostopoulos - 2020 - Edge-centric inferential modeling & analytics.pdf:application/pdf, 2020, August, Journal of Network and Computer Applications

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3