PPFL: A Parameter Behavior-Driven Plug-in Personalization Engine for Federated Learning
Qianyue Cao, Zongwei Zhu, Zirui Lian, Rui Zhang, Boyu Li, Yi Xiong, Xuehai Zhou
Abstract
Personalized Federated Learning (PFL) customizes models for each client to mitigate challenges from non-IID data, wherein a dominant strategy is model decoupling that partitions models into shared and personalized parts based on architectural priors (e.g., backbone vs. head). However, we reveal a critical flaw in this strategy: it induces "intrinsic drift," a performance degradation often more severe than the well-known client drift, which limits final accuracy. We trace this drift to a steep cliff of high loss emerging from the naive stitching of shared and personalized parts. To address this, we shift from architectural partitioning to a parameter behavior-driven paradigm. We introduce PPFL, an approach that employs a novel soft-fusion strategy guided by parameter-wise behavioral perception. PPFL dynamically infers each parameter
BibTeX
@inproceedings{aaai2026_ppflaparameterbe,
title = {PPFL: A Parameter Behavior-Driven Plug-in Personalization Engine for Federated Learning},
author = {Qianyue Cao and Zongwei Zhu and Zirui Lian and Rui Zhang and Boyu Li and Yi Xiong and Xuehai Zhou},
booktitle = {AAAI 2026},
year = {2026}
}