Poster: FedBlockParadox -- A Framework for Simulating and Securing Decentralized Federated Learning

2025-06-03

AI Summary

Existing decentralized federated learning (DFL) frameworks lack flexible, configurable simulation tools for comprehensive evaluation of diverse adversarial attacks. This paper introduces the first modular, open-source framework that unifies modeling of blockchain consensus protocols (e.g., PBFT, PoA), federated aggregation algorithms (e.g., FedAvg, Krum), and multiple attack classes (Byzantine, data poisoning, Sybil, etc.), while supporting configurable attack injection and dynamic validation mechanisms. Its key contributions include: (i) the first integration of multiple consensus protocols within a DFL simulation environment; (ii) fine-grained, parameterized adversarial modeling enabling precise robustness and security quantification; and (iii) empirical validation across 12 distinct attack scenarios, significantly improving efficiency and reproducibility of robustness analysis. The framework bridges a critical methodological gap in DFL security evaluation. Its source code is publicly available and has been adopted by multiple research groups for secure protocol verification and novel mechanism design.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems
• Security and Privacy: Blockchains and distributed ledgers

Abstract

A significant body of research in decentralized federated learning focuses on combining the privacy-preserving properties of federated learning with the resilience and transparency offered by blockchain-based systems. While these approaches are promising, they often lack flexible tools to evaluate system robustness under adversarial conditions. To fill this gap, we present FedBlockParadox, a modular framework for modeling and evaluating decentralized federated learning systems built on blockchain technologies, with a focus on resilience against a broad spectrum of adversarial attack scenarios. It supports multiple consensus protocols, validation methods, aggregation strategies, and configurable attack models. By enabling controlled experiments, FedBlockParadox provides a valuable resource for researchers developing secure, decentralized learning solutions. The framework is open-source and built to be extensible by the community.

Problem

• Simulating decentralized federated learning with blockchain resilience
• Evaluating system robustness under adversarial attack scenarios
• Providing modular tools for secure decentralized learning solutions

Innovation

• Modular framework for decentralized federated learning
• Supports multiple consensus and attack models
• Open-source and extensible for community development