FedBN: Federated Learning on Non-IID Features via Local Batch Normalization
read the original abstract
The emerging paradigm of federated learning (FL) strives to enable collaborative training of deep models on the network edge without centrally aggregating raw data and hence improving data privacy. In most cases, the assumption of independent and identically distributed samples across local clients does not hold for federated learning setups. Under this setting, neural network training performance may vary significantly according to the data distribution and even hurt training convergence. Most of the previous work has focused on a difference in the distribution of labels or client shifts. Unlike those settings, we address an important problem of FL, e.g., different scanners/sensors in medical imaging, different scenery distribution in autonomous driving (highway vs. city), where local clients store examples with different distributions compared to other clients, which we denote as feature shift non-iid. In this work, we propose an effective method that uses local batch normalization to alleviate the feature shift before averaging models. The resulting scheme, called FedBN, outperforms both classical FedAvg, as well as the state-of-the-art for non-iid data (FedProx) on our extensive experiments. These empirical results are supported by a convergence analysis that shows in a simplified setting that FedBN has a faster convergence rate than FedAvg. Code is available at https://github.com/med-air/FedBN.
This paper has not been read by Pith yet.
Forward citations
Cited by 12 Pith papers
-
Demystifying the Optimal Fair Classifier in Multi-Class Classification
Derives tractable optimal fair multi-class classifier and supplies in-processing and post-processing algorithms that converge to the accuracy-fairness Pareto frontier.
-
Personalized Digital Health Modeling with Adaptive Support Users
A new framework trains personal digital health models using adaptive weights on support users including dissimilar ones, achieving up to 25% lower RMSE in low-data settings.
-
Personalized Digital Health Modeling with Adaptive Support Users
A framework for personalizing digital health models by jointly learning adaptive weights on support users, using similarity-weighted transfer from similar users and contrastive regularization from dissimilar users to ...
-
When To Adapt? Adapting the Model or Data in Federated Medical Imaging
Harmonization works better than personalization for appearance-based domain shifts in federated medical imaging while personalization is superior for structural shifts, with both performing similarly when shifts are small.
-
FedSPDnet: Geometry-Aware Federated Deep Learning with SPDnet
FedSPDnet uses manifold projections and retractions to average Stiefel-constrained parameters in federated SPDnet, outperforming standard federated EEGnet on EEG motor imagery benchmarks in F1 score and robustness.
-
Bi-level Heterogeneous Learning for Time Series Foundation Models: A Federated Learning Approach
A bi-level federated learning framework trains time series foundation models on heterogeneous data by enforcing domain-invariant representations locally and improving cross-domain collaboration through aware aggregation.
-
FedPLT: Scalable, Resource-Efficient, and Heterogeneity-Aware Federated Learning via Partial Layer Training
FedPLT assigns client-specific model layers for training and matches or beats full-model federated learning accuracy with 71-82 percent fewer trainable parameters per client.
-
Domain-Adaptive Model Merging Across Disconnected Modes
DMM merges highly divergent domain-specific models without data sharing by synthesizing pseudo-data from normalization statistics and distilling knowledge, achieving state-of-the-art performance on unimodal and multim...
-
Benchmarking Federated Learning and Knowledge Distillation for Point Cloud Classification
Benchmark of federated learning plus knowledge distillation for point cloud classification reveals that label-free distillation objectives are required for student accuracy to reflect federated teacher quality rather ...
-
DP-FlogTinyLLM: Differentially private federated log anomaly detection using Tiny LLMs
DP-FLogTinyLLM combines federated learning, differential privacy, and LoRA-tuned tiny LLMs to match centralized log anomaly detection performance on Thunderbird and BGL datasets while preserving privacy.
-
Federated Breast Cancer Detection Enhanced by Synthetic Ultrasound Image Augmentation
Balanced synthetic image augmentation via GANs and diffusion models raises average AUC from 0.9206 to 0.9362 for FedAvg and 0.9429 to 0.9574 for FedProx in federated breast ultrasound classification.
-
Federated Weather Modeling on Sensor Data
A federated learning framework lets distributed weather sensors train shared deep learning models for forecasting and anomaly detection while keeping raw data private.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.