Bayesian Federated Learning

Bayesian Federated Learning (BFL) aims to leverage the power of distributed datasets for training machine learning models while preserving data privacy and quantifying prediction uncertainty. Current research focuses on developing efficient algorithms, such as those based on Hamiltonian Monte Carlo and variational inference, to address the communication overhead inherent in exchanging probabilistic model representations across multiple devices. This approach is particularly valuable in applications requiring reliable uncertainty estimates, such as those in industrial IoT and safety-critical systems, offering improvements over traditional frequentist federated learning methods. The resulting well-calibrated models and reduced communication costs are significant advancements for both theoretical understanding and practical deployment of machine learning in decentralized environments.

Papers

July 9, 2024

Bayesian Federated Learning with Hamiltonian Monte Carlo: Algorithm and Theory
Jiajun Liang, Qian Zhang, Wei Deng, Qifan Song, Guang Lin
Practical Algorithm Theoretical Understanding Hessian Matrix Convergence Guarantee Hamiltonian Monte Carlo Bayesian Federated Learning

June 4, 2024

One-Shot Federated Learning with Bayesian Pseudocoresets
Tim d'Hondt, Mykola Pechenizkiy, Robert Peharz
Approximate Inference Bayesian Federated Learning Function Space Variational Inference Bayesian Pseudocoreset

May 9, 2024

Compressed Bayesian Federated Learning for Reliable Passive Radio Sensing in Industrial IoT
Luca Barbieri, Stefano Savazzi, Monica Nicoli
Federated Learning Calibrated Learning Bayesian Federated Learning

March 22, 2024

Federated Bayesian Deep Learning: The Application of Statistical Aggregation Methods to Bayesian Models
John Fischer, Marko Orescanin, Justin Loomis, Patrick McClure
Application Proficiency Variational Inference Bayesian Deep Learning Bayesian Model Modern Deep Learning Mean Aggregation Bayesian Federated Learning

June 13, 2023

Privacy Preserving Bayesian Federated Learning in Heterogeneous Settings
Disha Makhija, Joydeep Ghosh, Nhat Ho
Privacy Policy Differential Privacy Guarantee Heterogeneous Setting Privacy Constraint Bayesian Federated Learning

April 26, 2023

Bayesian Federated Learning: A Survey
Longbing Cao, Hui Chen, Xuhui Fan, Joao Gama, Yew-Soon Ong, Vipin Kumar
Timely Survey Federated Learning Bayesian Perspective Bayesian Learning Bayesian Federated Learning

March 8, 2023

Federated Learning via Variational Bayesian Inference: Personalization, Sparsity and Clustering
Xu Zhang, Wenpeng Li, Yunfeng Shao, Yinchuan Li
Sparsity Increase Generalization Error Bound Variational Bayesian Inference Bayesian Federated Learning

September 14, 2022

Compressed Particle-Based Federated Bayesian Learning and Unlearning
Jinu Gong, Osvaldo Simeone, Joonhyuk Kang
Harmful Unlearning Communication Efficient Particle Based Variational Inference Uncertainty Information FL Algorithm Bayesian Federated Learning

February 24, 2022

Partitioned Variational Inference: A Framework for Probabilistic Federated Learning
Matthew Ashman, Thang D. Bui, Cuong V. Nguyen, Stratis Markou, Adrian Weller, Siddharth Swaroop, Richard E. Turner
New Framework Variational Inference Probabilistic Model Model Uncertainty Probability Estimation Bayesian Federated Learning