# Self-Aware Personalized Federated Learning

Source: [Self-Aware Personalized Federated Learning](<https://www.sota2.com/research/paper/68fa6c5c1e209131cea3395a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-24T17:14:13.669Z

Authors: Huili Chen, Jie Ding, Eric Tramel, Shuang Wu, Anit Kumar Sahu, Salman Avestimehr, Tao Zhang

Published: 2022-04-17T19:02:25.000Z

## Abstract

In the context of personalized federated learning (FL), the critical challenge is to balance local model improvement and global model tuning when the personal and global objectives may not be exactly aligned. Inspired by Bayesian hierarchical models, we develop a self-aware personalized FL method where each client can automatically balance the training of its local personal model and the global model that implicitly contributes to other clients' training. Such a balance is derived from the inter-client and intra-client uncertainty quantification. A larger inter-client variation implies more personalization is needed. Correspondingly, our method uses uncertainty-driven local training steps and aggregation rule instead of conventional local fine-tuning and sample size-based aggregation. With experimental studies on synthetic data, Amazon Alexa audio data, and public datasets such as MNIST, FEMNIST, CIFAR10, and Sent140, we show that our proposed method can achieve significantly improved personalization performance compared with the existing counterparts.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Wake-word detection on Wake-word detection dataset (test)](<https://www.sota2.com/research/sota/wake-word-detection-on-wake-word-detection-dataset-test.md>): Wake-word detection; Wake-word detection dataset (test); Relative FA (Device A): 0.92.

- [Personalized Federated Learning on FEMNIST (test)](<https://www.sota2.com/research/sota/personalized-federated-learning-on-femnist-test.md>): Personalized Federated Learning; FEMNIST (test); Top 10% Most Samples Accuracy: 76.24.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2204.08069>)

[PDF](<https://arxiv.org/pdf/2204.08069.pdf>)