FedCal: Achieving Local and Global Calibration in Federated Learning via Aggregated Parameterized Scaler

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Peng, Hongyi, Yu, Han, Tang, Xiaoli, Li, Xiaoxiao
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916272392896512
author Peng, Hongyi
Yu, Han
Tang, Xiaoli
Li, Xiaoxiao
author_facet Peng, Hongyi
Yu, Han
Tang, Xiaoli
Li, Xiaoxiao
contents Federated learning (FL) enables collaborative machine learning across distributed data owners, but data heterogeneity poses a challenge for model calibration. While prior work focused on improving accuracy for non-iid data, calibration remains under-explored. This study reveals existing FL aggregation approaches lead to sub-optimal calibration, and theoretical analysis shows despite constraining variance in clients' label distributions, global calibration error is still asymptotically lower bounded. To address this, we propose a novel Federated Calibration (FedCal) approach, emphasizing both local and global calibration. It leverages client-specific scalers for local calibration to effectively correct output misalignment without sacrificing prediction accuracy. These scalers are then aggregated via weight averaging to generate a global scaler, minimizing the global calibration error. Extensive experiments demonstrate FedCal significantly outperforms the best-performing baseline, reducing global calibration error by 47.66% on average.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15458
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FedCal: Achieving Local and Global Calibration in Federated Learning via Aggregated Parameterized Scaler
Peng, Hongyi
Yu, Han
Tang, Xiaoli
Li, Xiaoxiao
Machine Learning
Distributed, Parallel, and Cluster Computing
Federated learning (FL) enables collaborative machine learning across distributed data owners, but data heterogeneity poses a challenge for model calibration. While prior work focused on improving accuracy for non-iid data, calibration remains under-explored. This study reveals existing FL aggregation approaches lead to sub-optimal calibration, and theoretical analysis shows despite constraining variance in clients' label distributions, global calibration error is still asymptotically lower bounded. To address this, we propose a novel Federated Calibration (FedCal) approach, emphasizing both local and global calibration. It leverages client-specific scalers for local calibration to effectively correct output misalignment without sacrificing prediction accuracy. These scalers are then aggregated via weight averaging to generate a global scaler, minimizing the global calibration error. Extensive experiments demonstrate FedCal significantly outperforms the best-performing baseline, reducing global calibration error by 47.66% on average.
title FedCal: Achieving Local and Global Calibration in Federated Learning via Aggregated Parameterized Scaler
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2405.15458