Survey on federated learning threats: Concepts, taxonomy on attacks and defences, experimental study and challenges

Rodríguez-Barroso, Nuria; López, Daniel Jiménez; Luzón, M. Victoria; Herrera, Francisco; Martínez‐Cámara, Eugenio

doi:10.1016/j.inffus.2022.09.011

Cited by 114 publications

(33 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, varieties of attacks were proposed to access privacy risks in federated learning (FL) [21,28]. These attacks include threats such as model inversion [48], attribute inference [3], and membership inference.…”

Section: Attack Against Federated Learningmentioning

confidence: 99%

Interaction-level Membership Inference Attack Against Federated Recommender Systems

Wang¹,

Yang²,

Nguyen³

et al. 2023

Preprint

View full text Add to dashboard Cite

The marriage of federated learning and recommender system (Fe-dRec) has been widely used to address the growing data privacy concerns in personalized recommendation services. In FedRecs, users' attribute information and behavior data (i.e., user-item interaction data) are kept locally on their personal devices, therefore, it is considered a fairly secure approach to protect user privacy. As a result, the privacy issue of FedRecs is rarely explored. Unfortunately, several recent studies reveal that FedRecs are vulnerable to user attribute inference attacks, highlighting the privacy concerns of FedRecs. In this paper, we further investigate the privacy problem of user behavior data (i.e., user-item interactions) in FedRecs. Specifically, we perform the first systematic study on interactionlevel membership inference attacks on FedRecs. An interactionlevel membership inference attacker is first designed, and then the classical privacy protection mechanism, Local Differential Privacy (LDP), is adopted to defend against the membership inference attack. Unfortunately, the empirical analysis shows that LDP is not effective against such new attacks unless the recommendation performance is largely compromised. To mitigate the interactionlevel membership attack threats, we design a simple yet effective defense method to significantly reduce the attacker's inference accuracy without losing recommendation performance. Extensive experiments are conducted with two widely used FedRecs (Fed-NCF and Fed-LightGCN) on three real-world recommendation datasets (MovieLens-100K, Steam-200K, and Amazon Cell Phone), and the experimental results show the effectiveness of our solutions. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Section: Attack Against Federated Learningmentioning

confidence: 99%

Interaction-level Membership Inference Attack Against Federated Recommender Systems

Wang¹,

Yang²,

Nguyen³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Adversarial attacks are one of the biggest challenges in FL due to the large number of attacks and the issues of building a defense against such attacks. Rodríguez‐Barroso et al (2023) proposed a classification for all types of adversarial attacks: attacks to the federated model and privacy attacks. One example of an attack on the federated model is the issue in HFL wherein the participants can send poisoned updates to the server.…”

Section: Research Challenges and Future Directionsmentioning

confidence: 99%

Towards federated learning: An overview of methods and applications

Silva

Vinagre

Gama

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Federated learning (FL) is a collaborative, decentralized privacy-preserving method to attach the challenges of storing data and data privacy. Artificial intelligence, machine learning, smart devices, and deep learning have strongly marked the last years. Two challenges arose in data science as a result. First, the regulation protected the data by creating the General Data Protection Regulation, in which organizations are not allowed to keep or transfer data without the owner's authorization. Another challenge is the large volume of data generated in the era of big data, and keeping that data in one only server becomes increasingly tricky. Therefore, the data is allocated into different locations or generated by devices, creating the need to build models or perform calculations without transferring data to a single location. The new term FL emerged as a sub-area of machine learning that aims to solve the challenge of making distributed models with privacy considerations. This survey starts by describing relevant concepts, definitions, and methods, followed by an indepth investigation of federated model evaluation. Finally, we discuss three promising applications for further research: anomaly detection, distributed data streams, and graph representation.

show abstract

“…A privacy-preservation framework must ensure high protection to let individuals share their information. FL represents the most employed technology to accomplish the privacy task [ 3 , 4 , 32 ]. This federated technique facilitates distributed collaborative learning by multiple clients under the coordination of a server.…”

Section: State Of the Artmentioning

confidence: 99%

A Federated Learning-Inspired Evolutionary Algorithm: Application to Glucose Prediction

Falco

Cioppa

Koutný

et al. 2023

Sensors

View full text Add to dashboard Cite

In this paper, we propose an innovative Federated Learning-inspired evolutionary framework. Its main novelty is that this is the first time that an Evolutionary Algorithm is employed on its own to directly perform Federated Learning activity. A further novelty resides in the fact that, differently from the other Federated Learning frameworks in the literature, ours can efficiently deal at the same time with two relevant issues in Machine Learning, i.e., data privacy and interpretability of the solutions. Our framework consists of a master/slave approach in which each slave contains local data, protecting sensible private data, and exploits an evolutionary algorithm to generate prediction models. The master shares through the slaves the locally learned models that emerge on each slave. Sharing these local models results in global models. Being that data privacy and interpretability are very significant in the medical domain, the algorithm is tested to forecast future glucose values for diabetic patients by exploiting a Grammatical Evolution algorithm. The effectiveness of this knowledge-sharing process is assessed experimentally by comparing the proposed framework with another where no exchange of local models occurs. The results show that the performance of the proposed approach is better and demonstrate the validity of its sharing process for the emergence of local models for personal diabetes management, usable as efficient global models. When further subjects not involved in the learning process are considered, the models discovered by our framework show higher generalization capability than those achieved without knowledge sharing: the improvement provided by knowledge sharing is equal to about 3.03% for precision, 1.56% for recall, 3.17% for F1, and 1.56% for accuracy. Moreover, statistical analysis reveals the statistical superiority of model exchange with respect to the case of no exchange taking place.

show abstract

Survey on federated learning threats: Concepts, taxonomy on attacks and defences, experimental study and challenges

Cited by 114 publications

References 51 publications

Interaction-level Membership Inference Attack Against Federated Recommender Systems

Interaction-level Membership Inference Attack Against Federated Recommender Systems

Towards federated learning: An overview of methods and applications

A Federated Learning-Inspired Evolutionary Algorithm: Application to Glucose Prediction

Contact Info

Product

Resources

About