TPFL: A Trustworthy Personalized Federated Learning Framework via Subjective Logic

TL;DR

This paper introduces TPFL, a trustworthy federated learning framework using subjective logic to enhance model reliability, robustness, and safety in privacy-preserving collaborative classification tasks.

Contribution

The paper proposes a novel TPFL framework that employs subjective logic for probabilistic decision-making and uncertainty estimation, improving trustworthiness in federated learning.

Findings

Achieves competitive accuracy on federated benchmarks.

Demonstrates robustness against malicious attacks.

Ensures reliable performance under domain shifts.

Abstract

Federated learning (FL) enables collaborative model training across distributed clients while preserving data privacy. Despite its widespread adoption, most FL approaches focusing solely on privacy protection fall short in scenarios where trustworthiness is crucial, necessitating advancements in secure training, dependable decision-making mechanisms, robustness on corruptions, and enhanced performance with Non-IID data. To bridge this gap, we introduce Trustworthy Personalized Federated Learning (TPFL) framework designed for classification tasks via subjective logic in this paper. Specifically, TPFL adopts a unique approach by employing subjective logic to construct federated models, providing probabilistic decisions coupled with an assessment of uncertainty rather than mere probability assignments. By incorporating a trainable heterogeneity prior to the local training phase, TPFL effectively mitigates the adverse effects of data heterogeneity. Model uncertainty and instance uncertainty are further utilized to ensure the safety and reliability of the training and inference stages. Through extensive experiments on widely recognized federated learning benchmarks, we demonstrate that TPFL not only achieves competitive performance compared with advanced methods but also exhibits resilience against prevalent malicious attacks, robustness on domain shifts, and reliability in high-stake scenarios.