AdaBFL: Multi-Layer Defensive Adaptive Aggregation for Bzantine-Robust Federated Learning

arXiv cs.LG / 5/1/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces AdaBFL, a Byzantine-robust federated learning method designed to defend against poisoning attacks by malicious clients submitting corrupted updates.
AdaBFL uses a novel three-layer defensive aggregation scheme that adaptively re-weights defense algorithms to handle multiple, complex attack types.
It provides theoretical convergence guarantees for AdaBFL in a non-convex setting under non-IID data conditions.
Experiments on multiple datasets show AdaBFL outperforms comparable Byzantine-robust federated learning approaches.
The approach aims to improve robustness without relying on the server having access to the clients’ datasets, addressing a key limitation of prior methods.

Abstract

Federated learning (FL) is a popular distributed learning paradigm in machine learning, which enables multiple clients to collaboratively train models under the guidance of a server without exposing private client data. However, FL's decentralized nature makes it vulnerable to poisoning attacks, where malicious clients can submit corrupted models to manipulate the system. To counter such attacks, although various Byzantine-robust methods have been proposed, these methods struggle to provide balanced defense against multiple types of attacks or rely on possessing the dataset in the server. To deal with these drawbacks, thus, we propose an effective multi-layer defensive adaptive aggregation for Bzantine-robust federated learning (AdaBFL) based on a novel three-layer defensive mechanism, which can adaptively adjust the weights of defense algorithms to counter complex attacks. Moreover, we provide convergence properties of our AdaBFL method under the non-convex setting on non-iid data. Comprehensive experiments across multiple datasets validate the superiority of our AdaBFL over the comparable algorithms.