mlsanityData to 7 Oct 2026

Client Selection for Federated Bayesian Learning

arXiv 2212.05492 ↗First posted 11 Dec 2022cs.LG eess.SPcited by 4

Not on the map. Why?

Abstract (arXiv)

Distributed Stein Variational Gradient Descent (DSVGD) is a non-parametric distributed learning framework for federated Bayesian learning, where multiple clients jointly train a machine learning model by communicating a number of non-random and interacting particles with the server. Since communication resources are limited, selecting the clients with most informative local learning updates can improve the model convergence and communication efficiency. In this paper, we propose two selection schemes for DSVGD based on Kernelized Stein Discrepancy (KSD) and Hilbert Inner Product (HIP). We derive the upper bound on the decrease of the global free energy per iteration for both schemes, which is then minimized to speed up the model convergence. We evaluate and compare our schemes with conventional schemes in terms of model accuracy, convergence speed, and stability using various learning tasks and datasets.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 27 of its 44 references that are in the map. First posted on arXiv 11 Dec 2022.

Builds on

27 of 44 references.

References · 27

  1. 2017On Calibration of Modern Neural Networkscited by 4,860
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  3. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,531
  4. 2019On the Convergence of FedAvg on Non-IID Datacited by 1,220
  5. 2018Federated Learning with Non-IID Datacited by 1,099
  6. 2016Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithmcited by 498
  7. 2017Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Trainingcited by 471
  8. 2019Federated Machine Learning: Concept and Applicationscited by 409
  9. 2018Adaptive Federated Learning in Resource Constrained Edge Computing Systemscited by 408
  10. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  11. 2016A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluationcited by 236
  12. 2018Federated Learning via Over-the-Air Computationcited by 215
15 more
  1. 2018Broadband Analog Aggregation for Low-Latency Federated Edge Learning (Extended Version)cited by 186
  2. 2020One-Bit Over-the-Air Aggregation for Communication-Efficient Federated Edge Learning: Design and Convergence Analysiscited by 108
  3. 2014Stochastic Optimization with Importance Samplingcited by 91
  4. 2019Variational Federated Multi-Task Learningcited by 69
  5. 2020Scheduling for Cellular Federated Edge Learning with Importance and Channel Awarenesscited by 62
  6. 2020Fast-Convergent Federated Learningcited by 61
  7. 2018Sparse Binary Compression: Towards Distributed Deep Learning with minimal Communicationcited by 51
  8. 2019Accelerating DNN Training in Wireless Federated Edge Learning Systemscited by 34
  9. 2020Federated Generalized Bayesian Learning via Distributed Stein Variational Gradient Descentcited by 34
  10. 2018Partitioned Variational Inference: A unified framework encompassing federated and continual learningcited by 29
  11. 2016Patterns of Scalable Bayesian Inferencecited by 24
  12. 2017A Brief Introduction to Machine Learning for Engineerscited by 20
  13. 2019Data-Importance Aware User Scheduling for Communication-Efficient Edge Machine Learningcited by 20
  14. 2019Embarrassingly parallel MCMC using deep invertible transformationscited by 11
  15. 2020Data-Importance Aware Radio Resource Allocation: Wireless Communication Helps Machine Learningcited by 3

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 1; 2025: 1.

4 papers.

  1. 2023Accelerating Split Federated Learning over Wireless Communication Networkscited by 16
  2. 2023Asynchronous Wireless Federated Learning with Probabilistic Client Selectioncited by 4
  3. 2024SCALA: Split Federated Learning with Concatenated Activations and Logit Adjustmentscited by 2
  4. 2025A Two-Timescale Approach for Wireless Federated Learning with Parameter Freezing and Power Controlcited by 0