Towards Byzantine-Resilient Federated Learning via Group-Wise Robust Aggregation

Lei, Yu; Wu, Lingfei

doi:10.1007/978-3-030-63076-8_6

Cited by 11 publications

(8 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aggregation algorithm in FL is vulnerable to adversarial attacks [71], [72]. In the presence of label-flipping, backdoor attacks, and noisy updates, non-robust aggregation algorithms will produce unusable and compromised models.…”

Section: ) Non-robust Aggregationmentioning

confidence: 99%

Vulnerabilities in Federated Learning

Bouacida

Mohapatra

2021

IEEE Access

View full text Add to dashboard Cite

With more regulations tackling the protection of users' privacy-sensitive data in recent years, access to such data has become increasingly restricted. A new decentralized training paradigm, known as Federated Learning (FL), enables multiple clients located at different geographical locations to learn a machine learning model collaboratively without sharing their data. While FL has recently emerged as a promising solution to preserve users' privacy, this new paradigm's potential security implications may hinder its widespread adoption. The existing FL protocols exhibit new unique vulnerabilities that adversaries can exploit to compromise the trained model. FL is often preferred in learning environments where security and privacy are the key concerns. Therefore, it is crucial to raise awareness of the consequences resulting from the new threats to FL systems. To date, the security of traditional machine learning systems has been widely examined. However, many open challenges and complex questions are still surrounding FL security. In this paper, we bridge the gap in FL literature by providing a comprehensive survey of the unique security vulnerabilities exposed by the FL ecosystem. We highlight the vulnerabilities sources, key attacks on FL, defenses, as well as their unique challenges, and discuss promising future research directions towards more robust FL.INDEX TERMS Attacks, defenses, Federated Learning, security threats, vulnerabilities.

show abstract

Section: ) Non-robust Aggregationmentioning

confidence: 99%

Vulnerabilities in Federated Learning

Bouacida

Mohapatra

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Their work, however, is limited in that they considered only IID settings. [32] proposed a group-wise aggregation approach to address data heterogeneity, but not to defend against attacks. They developed a clustering algorithm on model parameters to group them so that if a new client comes in, its cluster assignment is determined by estimating the average center of the cluster.…”

Section: Defenses Against Anomalous Attacks In Federated Learningmentioning

confidence: 99%

ABC-FL: Anomalous and Benign client Classification in Federated Learning

Jeong¹,

Joonyong²,

Chung³

2021

Preprint

View full text Add to dashboard Cite

Federated Learning is a distributed machine learning framework designed for data privacy preservation i.e., local data remain private throughout the entire training and testing procedure. Federated Learning is gaining popularity because it allows one to use machine learning techniques while preserving privacy. However, it inherits the vulnerabilities and susceptibilities raised in deep learning techniques. For instance, Federated Learning is particularly vulnerable to data poisoning attacks that may deteriorate its performance and integrity due to its distributed nature and inaccessibility to the raw data. In addition, it is extremely difficult to correctly identify malicious clients due to the non-Independently and/or Identically Distributed (non-IID) data. The real-world data can be complex and diverse, making them hardly distinguishable from the malicious data without direct access to the raw data. Prior research has focused on detecting malicious clients while treating only the clients having IID data as benign. In this study, we propose a method that detects and classifies anomalous clients from benign clients when benign ones have non-IID data. Our proposed method leverages feature dimension reduction, dynamic clustering, and cosine similarity-based clipping. The experimental results validates that our proposed method not only classifies the malicious clients but also alleviates their negative influences from the entire procedure. Our findings may be used in future studies to effectively eliminate anomalous clients when building a model with diverse data. CCS CONCEPTS• Security and privacy; • Computing methodologies → Machine learning; Distributed artificial intelligence;

show abstract

“…However, the server end may not own such data and it is difficult to determine which data should be collected. Besides, Li et al [28] proposed to use conv layers' weights of uploaded models to build an auto-encoder to detect malicious clients and Yu & Wu [59] attempted to directly use the weights to distinguish malicious clients for robust aggregation. Fu et al [9] applied a weighting scheme to give a low weight to uploaded attacked models, mitigating their negative influences.…”

Section: Mobile Federated Learningmentioning

confidence: 99%

“…• IRLS [9]: IRLS applies a weighting scheme that gives a lower weight to uploaded attacked models, thus mitigating their negative influence. • GRA [59]: The goal of GRA is to directly use the weights to distinguish malicious clients for robust aggregation.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…For example, some researchers propose to achieve personalization by manually conducting a loss-based clustering [12] or a local adaptation after the FL process [29], which can improve the accuracy of traditional FL on non-iid data. People also propose to utilize model weights [28,59] or borrow extra data [23] to assist the detection of malicious clients and remove them. Unfortunately, existing techniques suffer from limitations such as requiring massive manual tuning or lacking of sufficient accuracy.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DistFL: Distribution-aware Federated Learning for Mobile Scenarios

Liu¹,

Cai²,

Zhang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Federated learning (FL) has emerged as an effective solution to decentralized and privacy-preserving machine learning for mobile clients. While traditional FL has demonstrated its superiority, it ignores the non-iid (independently identically distributed) situation, which widely exists in mobile scenarios. Failing to handle non-iid situations could cause problems such as performance decreasing and possible attacks. Previous studies focus on the "symptoms" directly, as they try to improve the accuracy or detect possible attacks by adding extra steps to conventional FL models. However, previous techniques overlook the root causes for the "symptoms": blindly aggregating models with the non-iid distributions. In this paper, we try to fundamentally address the issue by decomposing the overall non-iid situation into several iid clusters and conducting aggregation in each cluster. Specifically, we propose DistFL, a novel framework to achieve automated and accurate Distribution-aware Federated Learning in a cost-efficient way. DistFL achieves clustering via extracting and comparing the distribution knowledge from the uploaded models. With this framework, we are able to generate multiple personalized models with distinctive distributions and assign them to the corresponding clients. Extensive experiments on mobile scenarios with popular model architectures have demonstrated the effectiveness of DistFL.CCS Concepts: • Human-centered computing → Ubiquitous and mobile computing systems and tools; • Computing methodologies → Neural networks.

show abstract

Towards Byzantine-Resilient Federated Learning via Group-Wise Robust Aggregation

Cited by 11 publications

References 1 publication

Vulnerabilities in Federated Learning

Vulnerabilities in Federated Learning

ABC-FL: Anomalous and Benign client Classification in Federated Learning

DistFL: Distribution-aware Federated Learning for Mobile Scenarios

Contact Info

Product

Resources

About