Practical defences against model inversion attacks for split neural networks

Titcombe, Tom; Hall, Adam James; Papadopoulos, Pavlos; Romanini, Daniele

Abstract

We describe a threat model under which a split network-based federated learning system is susceptible to a model inversion attack by a malicious computational server. We demonstrate that the attack can be successfully performed with limited knowledge of the data distribution by the attacker. We propose a simple additive noise method to defend against model inversion, finding that the method can significantly reduce attack efficacy at an acceptable accuracy trade-off on MNIST. Furthermore, we show that NoPeekNN, an existing defensive method, protects different information from exposure, suggesting that a combined defence is necessary to fully protect private user data.

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: ICLR 2021 Workshop on Distributed and Private Machine Learning (DPML 2021)
• Start Date: May 7, 2021
• Publication Date: Apr 21, 2021
• Deposit Date: Oct 31, 2022
• Publicly Available Date: Nov 1, 2022
• Public URL: http://researchrepository.napier.ac.uk/Output/2946016
• Publisher URL: https://dp-ml.github.io/2021-workshop-ICLR/