A Novel Approach for Conducting Asynchronous Federated Learning with Blockchain Integration
Abstract
Abstract Conducting privacy-preserving collaborative learning over vertically partitioned data has emerged as a promising solution for multiparty joint modeling tasks. In such settings, data holders—including government agencies, financial institutions, and e-commerce platforms—can collaboratively train models without sharing their raw data or relying on trusted third parties. However, most existing vertical federated learning (VFL) approaches are constrained to synchronous training processes, which limits the efficiency of systems under heterogeneous computing and communication conditions. To address this limitation, an asynchronous VFL framework that improves the efficiency of training while preserving data privacy is proposed in this paper. To further provide enhanced security, trust, and participant engagement, blockchain technology is integrated into the proposed framework. Specifically, blockchain consensus mechanisms and smart contracts are leveraged to provide decentralized coordination, transparent model evaluation, and incentive-driven collaboration capabilities. Unlike traditional VFL systems that rely on a centralized coordinator, the proposed approach replaces the central server with a blockchain network. Task rewards are encapsulated within smart contracts, enabling the automatic and fair distribution of incentives based on a model quality evaluation. Participants are rewarded according to their contributions to the training process, thereby encouraging active…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.