Federated Learning With Residual Seasonality Adaptation (FedRSA): Robust Personalized Building Energy Forecasting in Heterogeneous, Data-Constrained Environments

Accurate building energy forecasting remains a challenge in heterogeneous sensing environments, non-IID data distributions, and limited local observations compounded by pronounced seasonality. We propose Federated Learning with Residual Seasonality Adaptation (FedRSA), a personalized federated learn...

ver descrição completa

Detalhes bibliográficos

Autor principal:	Sungwook Ryu
Formato:	Artigo
Idioma:	Inglês
Publicado em:	IEEE 2026-01-01
coleção:	IEEE Access
Assuntos:	Few-shot learning heterogeneous sensing environment personalized federated learning (pFL) residual seasonality adaptation smart building energy management
Acesso em linha:	https://ieeexplore.ieee.org/document/11471537/
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!