Federated Learning With Differential Privacy: Algorithms and Performance Analysis
By Kang Wei, Jun Li, Ming Ding, Chuan Ma, Howard H. Yang, Farokhi Farhad, Shi Jin, Tony Q. S. Quek, H. Vincent Poor. IEEE Transactions on Information Forensics and Security, 2019
2,251 citations recorded by the source corpus.
Federated learning (FL), as a type of distributed machine learning, is capable of significantly preserving clients’ private data from being exposed to adversaries. Nevertheless, private information can still be divulged by analyzing uploaded parameters from clients, e.g., weights trained in deep neural networks. In this paper, to effectively prevent information leakage, we propose a novel framework based on the concept of differential privacy (DP), in which artificial noise is added to parameters at the clients’ side before aggregating, namely, noising before model aggregation FL (NbAFL). First, we prove that the NbAFL can satisfy DP under distinct protection levels by properly adapting different variances of artificial noise. Then we develop a theoretical convergence bound on the loss function of the trained FL model in the NbAFL. Specifically, the theoretical bound reveals the following three key properties: 1) there is a tradeoff between convergence performance and privacy protection levels, i.e., better convergence performance leads to a lower protection level; 2) given a fixed privacy protection level, increasing the number <inline-formula> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> of overall clients participating in FL can improve the convergence performance; and 3) there is an optimal number aggregation times (communication rounds) in terms of convergence performance for a given protection level. Furthermore, we propose a <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula>-client random scheduling strategy, where <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> (<inline-formula> <tex-math notation="LaTeX">$1\leq K< N$ </tex-math></inline-formula>) clients are randomly selected from the <inline-formula> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> overall clients to participate in each aggregation. We also develop a corresponding convergence bound for the loss function in this case and the <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula>-client random scheduling strategy also retains the above three properties. Moreover, we find that there is an optimal <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> that achieves the best convergence performance at a fixed privacy level. Evaluations demonstrate that our theoretical results are consistent with simulations, thereby facilitating the design of various privacy-preserving FL algorithms with different tradeoff requirements on convergence performance and privacy levels.
Subfields: Artificial Intelligence, Federated & Distributed Learning, Privacy & Security
View Paper: https://arxiv.org/abs/1911.00222
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Federated & Distributed Learning
- Amir Salman Avestimehr
- Jinhyun So
- Basak Guler
- Qiang Yang
- Mingzhe Chen
- Philip S. Yu
- Jiankun Hu
- Yanming Zhu
In Privacy & Security
- Basak Guler
- Amir Salman Avestimehr
- Jinhyun So
- Philip S. Yu
- Nguyen Binh Truong
- Hangyu Zhu
- Bo Li
- Aruna Prasad Seneviratne
Citations in the corpus
- Turbo-Aggregate: Breaking the Quadratic Aggregation Barrier in Secure Federated Learning (2020, background): “from the local datasets cannot be identified from the local updates sent to the server, by adding artificial noise to the local updates at the clients’ side [9], [28], [29].”
- Federated Learning in Edge Computing: A Systematic Survey (2022, methodology): “The authors of [202] analyzed FL using the DP approach and designed a client selection method to improve privacy for subscribers.”
- Privacy preservation in federated learning: An insightful survey from the GDPR perspective (2020, background): “rameters updates and aggregation between local nodes and a central coordination server are strengthened by privacypreserving and cryptography techniques, which enhance data security and privacy [48, 123, 14, 15, 96].”
- Federated Learning Meets Blockchain in Edge Computing: Opportunities and Challenges (2021, background): “In particular, a shared global model is learned using distributed local datasets under the control of a blockchain which also provides model verification in a decentralized manner.”
- Asynchronous Federated Learning on Heterogeneous Devices: A Survey (2021, background): “Given that the training process in federated learning (FL) typically involves numerous local devices, sometimes numbering in the dozens or even thousands, these concerns can be mitigated by setting the mean value of the Laplace distribution to zero [44, 45].”
- Trustworthy AI: From Principles to Practices (2021, methodology): “It has also been applied to single- and multi-machine computing environments, and integrated with ML models for protecting model privacy [2, 120, 349].”
- Federated Learning in Smart City Sensing: Challenges and Opportunities (2020, neutral): “The parameters of the local model are then transmitted to the central server, while the participating user’s local data remain on their devices [41].”
- A Comprehensive Survey of Privacy-preserving Federated Learning (2021, methodology): “WHO RA1 [61, 196, 228] [26, 137, 218] [157, 169, 198] [144, 164, 199] [63, 70, 193] [16, 174, 191] [78, 110, 228] [52, 81, 216] [32, 56, 222] [30, 127, 173] RA2 [114, 175, 187] [215, 221, 224] [29, 111, 113] [13, 199] [3, 10, 40] [109, 155, 168] [45, 172, 193] [20, 52, 57] [56, 165, 197] [173, 199, ”
- Federated Learning for Smart Healthcare: A Survey (2021, methodology): “A differential privacy-based solution is adopted in [82], where each”
- Federated Learning for 6G: Applications, Challenges, and Opportunities (2021, background): “Regarding the privacy protection for each user and the BS, a promising approach is to use differential privacy, which introduces a tradeoff between privacy and FL performance [93].”
- A Full Dive Into Realizing the Edge-Enabled Metaverse: Visions, Enabling Technologies, and Challenges (2022, background): “The authors in [230] add noise before sending the model parameters to the server for aggregation.”
- Federated Learning on Non-IID Data: A Survey (2021, methodology): “For instance, local DP is used in [120, 37, 167, 126, 114, 137], in which the model gradients on the client side are perturbed by adding Gaussian or Laplacian noise before uploading them to the server.”
- Decentralized Federated Learning: A Survey and Perspective (2023, background): “Wei et al. [81] took into account the privacy leakage concern of model parameters uploaded by clients in FL and proposed to improve the differential privacy by adding noise before the client sends it to the server for aggregation.”