Amplitude-Varying Perturbation for Balancing Privacy and Utility in Federated Learning
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
While preserving the privacy of federated learning (FL), differential privacy (DP) inevitably degrades the utility (i.e., accuracy) of FL due to model perturbations caused by DP noise added to model updates. Existing studies have considered exclusively noise with persistent root-mean-square amplitude and overlooked an opportunity of adjusting the amplitudes to alleviate the adverse effects of the noise. This paper presents a new DP perturbation mechanism with a time-varying noise amplitude to protect the privacy of FL and retain the capability of adjusting the learning performance. Specifically, we propose a geometric series form for the noise amplitude and reveal analytically the dependence of the series on the number of global aggregations and the (ϵ,δ)-DP requirement. We derive an online refinement of the series to prevent FL from premature convergence resulting from excessive perturbation noise. Another important aspect is an upper bound developed for the loss function of a multi-layer perceptron (MLP) trained by FL running the new DP mechanism. Accordingly, the optimal number of global aggregations is obtained, balancing the learning and privacy. Extensive experiments are conducted using MLP, supporting vector machine, and convolutional neural network models on four public datasets. The contribution of the new DP mechanism to the convergence and accuracy of privacy-preserving FL is corroborated, compared to the state-of-the-art Gaussian noise mechanism with a persistent noise amplitude.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Federated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322 · cited 1×
“It provides an effective means of model training without directly leaking private information [2].”
BackgroundFederated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210 · cited 1×
“3258255 any sensitive raw data of the devices to central servers [1].”
BackgroundThe Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557
Membership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754 · cited 1×
“[5] demonstrated that private information about local datasets can be derived from trained local models.”
BackgroundThe Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
Federated Learning With Differential Privacy: Algorithms and Performance Analysis
Wei, K., Li, J., Ding, M., Ma, C., Yang, H. H., Farokhi, F., Jin, S., Quek, T. Q. S., & Vincent Poor, H. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2988575 · Cited by 2,400 · cited 8×
“Accordingly, if the batch size for training the local model is consistent with the training sample number, the sensitivity is given by 1s = 2C |Dk | with C being the pre-determined clipping threshold and Dk being the local dataset at the k-th user [27].”
BackgroundMethodInfluentialSelf-citationAdaptive Federated Learning in Resource Constrained Edge Computing Systems
Wang, S., Tuor, T., Salonidis, T., Leung, K. K., Makaya, C., He, T., & Chan, K. (2019) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2019.2904348 · Cited by 2,232
Mironov, I. (2017) · 2017 IEEE 30th Computer Security Foundations Symposium (CSF) · doi:10.1109/csf.2017.11 · Cited by 1,255 · cited 1×
“Differential privacy (DP) [21] is the de facto privacy mechanism that has been increasingly studied, including ε-DP, (ε, δ)-DP, Rényi DP, and (α, ε)-Rényi DP [21], [22], [23], [24], [25].”
BackgroundLocal Differential Privacy-Based Federated Learning for Internet of Things
Zhao, Y., Zhao, J., Yang, M., Wang, T., Wang, N., Lyu, L., Niyato, D., & Lam, K.-Y. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3037194 · Cited by 456 · cited 2×
“, [26], [27], and [28], have considered time-invariant DP noise perturbations; in other words, the variance of the DP perturbation noise remains persistent among global aggregations.”
BackgroundLDP-Fed: federated learning with local differential privacy
Truex, S., Liu, L., Chow, K.-H., Gursoy, M. E., & Wei, W. (2020) · Proceedings of the Third ACM International Workshop on Edge Systems, Analytics and Networking · doi:10.1145/3378679.3394533 · Cited by 435 · cited 2×
“, [26], [27], and [28], have considered time-invariant DP noise perturbations; in other words, the variance of the DP perturbation noise remains persistent among global aggregations.”
Background
Cited by
Wan, Y., Qu, Y., Ni, W., Xiang, Y., Gao, L., & Hossain, E. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3361451 · Cited by 152
“To further protect participants’ private information when exchanging information during the training process [71], many encryption algorithms and authentication techniques for wireless networks have been proposed [72].”
BackgroundSelf-citationYou, F., Yuan, X., Ni, W., & Jamalipour, A. (2025) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3499342 · Cited by 15
“…δ ≥ 0 , a privacy-preserving mechanism M is ( (cid:9), δ ) -LDP if and only if, for any users j and j , users j and j ’s transitions X j , X j ∈ X , and all S ⊆ Range ( M ) , Pr M X j ∈ S ≤ exp( (cid:9) ) × Pr M X j ∈ S + δ, (21) where X j is the transition space associated with user j [25], [26].”
Self-citationA Novel Privacy-Preserving Incentive Mechanism for Multi-Access Edge Computing
You, F., Yuan, X., Ni, W., & Jamalipour, A. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3391303 · Cited by 11
Self-citation
Metadata
- Authors
- Yuan, X., Ni, W., Ding, M., Wei, K., Li, J., & Poor, H. V.
- Venue
- IEEE Transactions on Information Forensics and Security
- Published
- 2023
- Volume
- 18
- Pages
- 1884-1897
- ISSN
- 1556-6013 (ISSN-L)
- 1556-6021
- OpenAlex
- W4327643804
- Cited by
- 80
- References
- 13
- FWCI
- 7.03
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Amplitude-Varying Perturbation for Balancing Privacy and Utility in Federated Learning. https://app.cochat.ai/papers/10.1109/tifs.2023.3258255
- The paper
- Yuan, X., Ni, W., Ding, M., Wei, K., Li, J., & Poor, H. V. (2023). Amplitude-Varying Perturbation for Balancing Privacy and Utility in Federated Learning. IEEE Transactions on Information Forensics and Security. https://doi.org/10.1109/tifs.2023.3258255
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.