Split computing reliability and security combined for edge dnn robustness
JASPER: Special Session on Joint Reliability And Security Assessment of SPlit Computing for Edge Robustness
Cryptography and SecurityHardware ArchitectureDistributed, Parallel, and Cluster ComputingMachine Learning
Summary
Split Computing means splitting a neural network’s job between a nearby device and a cloud server. The authors noticed that the data passed between these two parts can get messed up either by hardware problems or by hackers, but usually these problems are studied separately. They created a new way to check both problems together using special scores from simulated faults and attacks on a well-known neural network. Their tests showed different ways to break the system and how looking at both problems together gives a better understanding of where it is vulnerable.
What this means in practice
- •For edge device developers: Assess combined hardware faults and security risks to design more robust edge-cloud DNN deployments.
- •For cloud infrastructure teams: Use joint vulnerability metrics to optimize split computing configurations for reliable and secure inference services.
Authors
Enrico Magliano, Giuseppe Esposito, Amir Hossein Shahdadian, Rama Mounika Kodamanchili, Juan David Guerrero Balaguera, Juan Esteban Rodriguez Condia, Annachiara Ruospo, Roberta Siciliano, Stefano Di Carlo, Maksim Jenihhin, Marco Levorato, Alessandro Savino, Matteo Sonza Reorda, Christian Herglotz, Michael Hübner, Mahdi Taheri
Abstract
Split Computing (SC) enables efficient deployment of Deep Neural Networks (DNNs) by partitioning inference between edge devices and cloud servers. However, intermediate feature representations are simultaneously exposed to hardware faults and adversarial attacks, which are traditionally evaluated independently. This paper presents a unified framework for the joint assessment of reliability and security in Split Computing. First, reliability is characterized through neuron-level fault injection using the Mean Relative Accuracy Degradation (MRAD) while security through feature-map-aware adversarial attacks simulations using the Attack Success Rate (ASR). Based on these complementary analyses, the Joint Vulnerability Score (JVS) is introduced, along with a confidence-aware extension that jointly captures prediction errors and confidence degradation. The framework is evaluated on ten Split Computing configurations based on ResNet-50 trained on ILSVRC-2012. Experimental results show substantial differences across compression strategies, with MRAD ranging from 44.3% to 61.2% under fault injection, while adversarial attacks achieve up to 98.8% ASR. Furthermore, the proposed joint metrics reveal vulnerability trends that remain hidden when reliability and security are analyzed independently, providing a more comprehensive methodology for designing dependable Split Computing systems.