mlsanityData to 14 Oct 2026

Leveraging Channel Noise for Sampling and Privacy via Quantized Federated Langevin Monte Carlo

arXiv 2202.13932 ↗First posted 28 Feb 2022cs.IT eess.SP math.ITcited by 3

Not on the map. Why?

Abstract (arXiv)

For engineering applications of artificial intelligence, Bayesian learning holds significant advantages over standard frequentist learning, including the capacity to quantify uncertainty. Langevin Monte Carlo (LMC) is an efficient gradient-based approximate Bayesian learning strategy that aims at producing samples drawn from the posterior distribution of the model parameters. Prior work focused on a distributed implementation of LMC over a multi-access wireless channel via analog modulation. In contrast, this paper proposes quantized federated LMC (FLMC), which integrates one-bit stochastic quantization of the local gradients with channel-driven sampling. Channel-driven sampling leverages channel noise for the purpose of contributing to Monte Carlo sampling, while also serving the role of privacy mechanism. Analog and digital implementations of wireless LMC are compared as a function of differential privacy (DP) requirements, revealing the advantages of the latter at sufficiently high signal-to-noise ratio.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 17 of its 24 references that are in the map. First posted on arXiv 28 Feb 2022.

Builds on

17 of 24 references.

References · 17

  1. 2019Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computingcited by 228
  2. 2020Understanding Gradient Clipping in Private SGD: A Geometric Perspectivecited by 142
  3. 2020One-Bit Over-the-Air Aggregation for Communication-Efficient Federated Edge Learning: Design and Convergence Analysiscited by 112
  4. 2021The Bayesian Learning Rulecited by 99
  5. 2018Towards an Intelligent Edge: Wireless Communication Meets Machine Learningcited by 95
  6. 2018Wireless Network Intelligence at the Edgecited by 92
  7. 2020Over-the-Air Federated Learning from Heterogeneous Datacited by 73
  8. 2015Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlocited by 63
  9. 2020Privacy For Free: Wireless Federated Learning Via Uncoded Transmission With Adaptive Power Controlcited by 63
  10. 2020Wireless Federated Learning with Local Differential Privacycited by 52
  11. 2021Optimized Power Control Design for Over-the-Air Federated Edge Learningcited by 48
  12. 2020Stochastic-Sign SGD for Federated Learning with Theoretical Guaranteescited by 35
5 more
  1. 2019Machine Learning in the Aircited by 33
  2. 2019Over-the-Air Computation Systems: Optimization, Analysis and Scaling Lawscited by 27
  3. 2016Patterns of Scalable Bayesian Inferencecited by 25
  4. 2020Differentially Private AirComp Federated Learning with Power Adaptation Harnessing Receiver Noisecited by 15
  5. 2021Wireless Federated Langevin Monte Carlo: Repurposing Channel Noise for Bayesian Sampling and Privacycited by 10

Built on by

Cited by year

Citing papers per year: 2022: 3.

3 papers.

  1. 2022Joint Privacy Enhancement and Quantization in Federated Learningcited by 35
  2. 2022A Survey on Over-the-Air Computationcited by 21
  3. 2022Federated Averaging Langevin Dynamics: Toward a unified theory and new algorithmscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2023Layered Randomized Quantization for Communication-Efficient and Privacy-Preserving Distributed Learning
  2. 2022Optimal Privacy Preserving for Federated Learning in Mobile Edge Computing
  3. 2023Killing Two Birds with One Stone: Quantization Achieves Privacy in Distributed Learning
  4. 2025Privacy-Preserving Quantized Federated Learning with Diverse Precision
  5. 2024The Effect of Quantization in Federated Learning: A R'enyi Differential Privacy Perspective
  6. 2026Quantization in Federated Learning: Methods, Challenges and Future Directions
  7. 2022Quantization enabled Privacy Protection in Decentralized Stochastic Optimization
  8. 2026Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy