No free lunch theorem for security and utility in federated learning

X Zhang, H Gu, L Fan, K Chen, Q Yang - ACM Transactions on Intelligent …, 2022 - dl.acm.org
… article may serve as the guidance for the design of practical federated learning algorithms. …
into account, we therefore reiterate that a secure federated learning (SFL) scheme must (1) …

Trading off privacy, utility, and efficiency in federated learning

X Zhang, Y Kang, K Chen, L Fan, Q Yang - ACM Transactions on …, 2023 - dl.acm.org
… a No-Free-Lunch (NFL) Theorem (Theorem 4.6) that provides a … Our No-Free-Lunch theorem
is formulated based on this … No-Free-Lunch theorem (Theorem 4.6) for federated learning, …

A theorem of the alternative for personalized federated learning

S Chen, Q Zheng, Q Long, WJ Su - arXiv preprint arXiv:2103.01901, 2021 - arxiv.org
… A widely recognized difficulty in federated learning arises from the … risks of personalized
federated learning with a smooth, … theorem of the alternative for personalized federated learning: …

A Game-theoretic Framework for Privacy-preserving Federated Learning

X Zhang, L Fan, S Wang, W Li, K Chen… - ACM Transactions on …, 2024 - dl.acm.org
… No free lunch theorem for security and utility in federated learning. ACM Transactions on
Intelligent Systems and Technology 14, 1 (2022), 1ś35. [44] Xiaojin Zhang, Anbu Huang, Lixin …

Theoretical Analysis of Privacy Leakage in Trustworthy Federated Learning: A Perspective from Linear Algebra and Optimization Theory

X Zhang, W Chen - arXiv preprint arXiv:2407.16735, 2024 - arxiv.org
… analysis of privacy leakage in federated learning from two perspectives: linear … federated
learning, offering a theoretical foundation for designing privacypreserving federated learning

No Free Lunch Theorem for Privacy-Preserving LLM Inference

X Zhang, Y Fei, Y Kang, W Chen, L Fan, H Jin… - arXiv preprint arXiv …, 2024 - arxiv.org
… -Free-Lunch Theorem. Another pertinent research [7] similarly introduces a theorem that
try … models in the context of horizontal federated learning, where privacy leakage is defined …

Fedeval: A holistic evaluation framework for federated learning

D Chai, L Wang, L Yang, J Zhang, K Chen… - arXiv preprint arXiv …, 2020 - arxiv.org
… In federated learning, we typically learn the global model by solving the following problem: …
Yang, “No free lunch theorem for security and utility in federated learning,” arXiv preprint arXiv…

The Aggregation–Heterogeneity Trade-off in Federated Learning

X Zhao, H Wang, W Lin - … Annual Conference on Learning …, 2023 - proceedings.mlr.press
learning holds that the more data you train your model on, the better the model can perform.
Accordingly, a plethora of federated learning … and heterogeneity in federated learning. We …

Attacks against federated learning defense systems and their mitigation

C Lewis, V Varadharajan, N Noman - Journal of Machine Learning …, 2023 - jmlr.org
… in federated learning defense systems (FLDS) and present a detailed empirical analysis
of these attacks under different conditions. (2) Then we propose a new federated learning

Server free wireless federated learning: Architecture, algorithm, and analysis

HH Yang, Z Chen, TQS Quek - arXiv preprint arXiv:2204.07609, 2022 - arxiv.org
… Abstract—We demonstrate that merely analog transmissions and match filtering can
realize the function of an edge server in federated learning (FL). Therefore, a network with …