Resilience of Wireless Ad Hoc Federated Learning against Model Poisoning Attacks

TL;DR

This paper analyzes the resilience of wireless ad hoc federated learning (WAFL) against model poisoning attacks, showing that WAFL can recover from attacks and even improve accuracy after attackers leave, due to its decentralized nature.

Contribution

The paper provides a theoretical analysis of WAFL's resilience to poisoning attacks and validates it through experiments demonstrating recovery and improved accuracy post-attack.

Findings

Nodes directly encountering attackers are compromised but others remain resilient.

After attackers leave, nodes recover and achieve higher accuracy than no-attack scenarios.

WAFL's decentralized approach enhances robustness against poisoning attacks.

Abstract

Wireless ad hoc federated learning (WAFL) is a fully decentralized collaborative machine learning framework organized by opportunistically encountered mobile nodes. Compared to conventional federated learning, WAFL performs model training by weakly synchronizing the model parameters with others, and this shows great resilience to a poisoned model injected by an attacker. In this paper, we provide our theoretical analysis of the WAFL's resilience against model poisoning attacks, by formulating the force balance between the poisoned model and the legitimate model. According to our experiments, we confirmed that the nodes directly encountered the attacker has been somehow compromised to the poisoned model but other nodes have shown great resilience. More importantly, after the attacker has left the network, all the nodes have finally found stronger model parameters combined with the poisoned model. Most of the attack-experienced cases achieved higher accuracy than the no-attack-experienced cases.