Analyzing User-Level Privacy Attack Against Federated Learning

Song, Mengkai; Wang, Zhen; Zhang, Zhifei; Song, Yang; Wang, Qian; Ren, Ju; Qi, Hairong

doi:10.1109/jsac.2020.3000372

Cited by 187 publications

(81 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To meet the constraints of accuracy, memory, and energy in IoT end devices, models with mixed-precision are constructed. The paper titled "Analyzing User-Level Privacy Attack Against Federated Learning," by Song et al, focuses on attacks by a malicious server [16]. The authors introduce an attack framework called mGAN-AI that is based on Generative Adversarial Networks (GANs).…”

Section: The Selected Papersmentioning

confidence: 99%

“…Many works apply deep learning methods, which are based on neural networks [2]- [5], [8]- [10], [13], [15], [16]. Some papers use methods based on Recurrent Neural Networks (RNNs), specifically Long Short-Term Memory (LSTM) [5], [8].…”

Section: The Selected Papersmentioning

confidence: 99%

“…Some papers use methods based on Recurrent Neural Networks (RNNs), specifically Long Short-Term Memory (LSTM) [5], [8]. One work uses a Graph Neural Network (GNN) to reflect relationships in networked systems [4], and one uses Generative Adversarial Networks (GANs) to address privacy issues [16].…”

Section: The Selected Papersmentioning

confidence: 99%

“…The paper [4] proposes a meta-learning scheme that allows to switch among sub-policies. Finally, [16] applies federated learning to build a solution for user privacy in case of malicious server.…”

Section: The Selected Papersmentioning

confidence: 99%

See 3 more Smart Citations

Guest Editorial Special Issue on Advances in Artificial Intelligence and Machine Learning for Networking

Chemouil

Hui

Kellerer

et al. 2020

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

I. INTRODUCTION A RTIFICIAL Intelligence (AI) and Machine Learning (ML) approaches have emerged in the networking domain with great expectation. They can be broadly divided into AI/ML techniques for network engineering and management, network designs for AI/ML applications, and system concepts. AI/ML techniques for networking and management improve the way we address networking. They support efficient, rapid, and trustworthy engineering, operations, and management. As such, they meet the current interest in softwarization and network programmability that fuels the need for improved network automation in agile infrastructures, including edge and fog environments. Network design and optimization for AI/ML applications addresses the complementary topic of supporting AI/ML-based systems through novel networking techniques, including new architectures and algorithms. The third topic area is system implementation and open-source software development. This evolution draws particular attention to interdisciplinary approaches. Researchers in communication networks apply ML and AI concepts to optimize and automate network architecture, control, and management. Similarly, AI experts collaborate with networking researchers to optimize network support for architecture and design of data communication and processing for AI purposes. This special issue is a follow-up to the JSAC's Special Issue on Artificial Intelligence and Machine Learning for Networking and Communications published in June 2019 [1]. It has been organized by the same core team of researchers.

show abstract

Section: The Selected Papersmentioning

confidence: 99%

Section: The Selected Papersmentioning

confidence: 99%

Section: The Selected Papersmentioning

confidence: 99%

Section: The Selected Papersmentioning

confidence: 99%

See 2 more Smart Citations

Guest Editorial Special Issue on Advances in Artificial Intelligence and Machine Learning for Networking

Chemouil

Hui

Kellerer

et al. 2020

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

show abstract

“…Privacy leakage is particularly serious, and many security researchers have put forward their own methods and opinions. For instance, Mengkai Song et al analyzed the problem of users' privacy being leaked due to malicious attacks on servers [3]. Vulnerability analysis of IoT devices is a prerequisite for reducing IoT security threats, and IoT devices identification technology is the predominate method of IoT devices vulnerability analysis.…”

Section: Introductionmentioning

confidence: 99%

Cross-Layer Protocol Fingerprint for Large-Scale Fine-Grain Devices Identification

Yu¹,

Xin²,

Chen³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Internet-connected Internet of Things (IoT) devices are exploding, which pose a significant threat for their management and security protection. IoT device identification is a prerequisite for discovering, monitoring, and protecting these devices. Although the existing proactive identification methods based on protocol fingerprint can discover and identify large-scale IoT devices, the fingerprint granularity is difficult to meet the requirements of security risk assessment for large-scale IoT devices. Since IoT devices usually support multiple network protocols for specific collection and control tasks, we propose a cross-layer protocol fingerprint to achieve large-scale fine-grained devices identification instead of traditional single protocol fingerprint. We first design a probing scheme for gathering HTTP and TCP cross-layer packets. Then we select the specific field of the HTTP and TCP protocols based on the diversity and consistence of field value. Finally, we utilize convolutional neural network (CNN) and long-term memory network (LSTM) to extract and construct feature fingerprint of these specific fields, and achieve a fine-grain IoT devices identification with high accuracy. The experimental results show that our identification accuracy of devices model reaches 96.6%, the recall rate reaches 97.4%.

show abstract

A federated learning method for network intrusion detection

Tang

2021

Concurrency and Computation

View full text Add to dashboard Cite

Intrusion detection is a common network security defense technology. At present, there are many research using deep learning to realize network intrusion detection. This method has been proved to have high detection accuracy. However, deep learning requires large-scale data sets for training. The network intrusion detection data set of some institution is lacking. If the network traffic data set is uploaded for centralized deep learning training, it will face privacy problems. Combined with the characteristics of network traffic, this article proposes a network intrusion detection method based on federated learning. This method allows multiple ISPs or other institutions to conduct joint deep learning training on the premise of retaining local data. It not only improves the detection accuracy of the model but also protects privacy in network traffic. This article conducts experiments on the CICIDS2017 network intrusion detection data set. Experimental results show that worker participating in federated learning have higher detection accuracy. The accuracy and other performance of federated learning are almost equal to those of centralized deep learning models.

show abstract

Analyzing User-Level Privacy Attack Against Federated Learning

Cited by 187 publications

References 23 publications

Guest Editorial Special Issue on Advances in Artificial Intelligence and Machine Learning for Networking

Guest Editorial Special Issue on Advances in Artificial Intelligence and Machine Learning for Networking

Cross-Layer Protocol Fingerprint for Large-Scale Fine-Grain Devices Identification

A federated learning method for network intrusion detection

Contact Info

Product

Resources

About