FedPS: Federated data Preprocessing via aggregated Statistics

2026-02-11

AI Summary

This work addresses the lack of standardized data preprocessing methods in federated learning that simultaneously ensure privacy preservation and communication efficiency, particularly in handling missing values, inconsistent data formats, and heterogeneous feature scales. To this end, we propose FedPS, the first unified federated preprocessing framework that supports feature scaling, encoding, discretization, and missing value imputation, compatible with both horizontal and vertical federated settings. By leveraging data sketching techniques to efficiently summarize local statistics and integrating secure aggregation protocols, FedPS enables privacy-preserving preprocessing with low communication overhead. Furthermore, the framework extends algorithms such as k-Means, kNN, and Bayesian linear regression to the federated paradigm, significantly enhancing model performance and practicality on real-world heterogeneous datasets while maintaining consistency across participants.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Natural Language Processing: Summarization
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Security and Privacy: Privacy-enhancing technologies

Abstract

Federated Learning (FL) enables multiple parties to collaboratively train machine learning models without sharing raw data. However, before training, data must be preprocessed to address missing values, inconsistent formats, and heterogeneous feature scales. This preprocessing stage is critical for model performance but is largely overlooked in FL research. In practical FL systems, privacy constraints prohibit centralizing raw data, while communication efficiency introduces further challenges for distributed preprocessing. We introduce FedPS, a unified framework for federated data preprocessing based on aggregated statistics. FedPS leverages data-sketching techniques to efficiently summarize local datasets while preserving essential statistical information. Building on these summaries, we design federated algorithms for feature scaling, encoding, discretization, and missing-value imputation, and extend preprocessing-related models such as k-Means, k-Nearest Neighbors, and Bayesian Linear Regression to both horizontal and vertical FL settings. FedPS provides flexible, communication-efficient, and consistent preprocessing pipelines for practical FL deployments.

Problem

• Federated Learning
• Data Preprocessing
• Privacy Preservation
• Communication Efficiency
• Statistical Aggregation

Innovation

• Federated Learning
• Data Preprocessing
• Aggregated Statistics
• Data Sketching
• Privacy-Preserving