%0 Conference Proceedings %T A Practical Introduction to Side-Channel Extraction of Deep Neural Network Parameters %+ Département Systèmes (DSYS) %A Joud, Raphaël %A Moellic, Pierre-Alain %A Pontie, Simon %A Rigaud, Jean-Baptiste %B CARDIS 2022 %C Birmingham, United Kingdom %3 Springer Lecture Notes in Computer Science (LNCS) %P 45-65 %8 2022-11-07 %D 2022 %Z 2211.05590 %R 10.1007/978-3-031-25319-5_3 %Z Computer Science [cs]/Artificial Intelligence [cs.AI] %Z Engineering Sciences [physics]/ElectronicsConference papers %X Model extraction is a major threat for embedded deep neural network models that leverages an extended attack surface. Indeed, by physically accessing a device, an adversary may exploit side-channel leakages to extract critical information of a model (i.e., its architecture or internal parameters). Different adversarial objectives are possible includinga fidelity-based scenario where the architecture and parameters are precisely extracted (model cloning). We focus this work on software implementation of deep neural networks embedded in a high-end 32-bit microcontroller (Cortex-M7) and expose several challenges related to fidelity-based parameters extraction through side-channel analysis, from the basic multiplication operation to the feed-forward connection through the layers. To precisely extract the value of parameters represented in the single-precision floating point IEEE-754 standard, we propose an iterative process that is evaluated with both simulations and traces from a Cortex-M7 target. To our knowledge, this work is the first to target such an high-end 32-bit platform. Importantly, we raise and discuss remaining challenges for the complete extraction of a deep neural network model, more particularly the critical case of biases. %G English %L cea-04038159 %U https://cea.hal.science/cea-04038159 %~ CEA