BoMaNet
Anuj Dubey, Rosario Cammarota, Aydın Aysu
Abstract
Recent work on stealing machine learning (ML) models from inference engines with physical side-channel attacks warrant an urgent need for effective side-channel defenses. This work proposes the first fully-masked neural network inference engine design.
Topics & Concepts
Computer scienceInferenceWarrantSide channel attackArtificial neural networkInference engineArtificial intelligenceMachine learningChannel (broadcasting)Work (physics)Computer securityEngineeringTelecommunicationsCryptographyMechanical engineeringFinancial economicsEconomicsAdversarial Robustness in Machine LearningCryptographic Implementations and SecurityAdvanced Malware Detection Techniques