arXiv subjects

Kasra Ahmadi

Publications and source records attributed to Kasra Ahmadi.

5 recordsLinked to original sources

arXiv2026

MAED: Mathematical Activation Error Detection for Mitigating Physical Fault Attacks in DNN Inference

The inference phase of deep neural networks (DNNs) in embedded systems is increasingly vulnerable to fault attacks and failures, which can result in incorrect predictions. These vulnerabilities can potentially lead to catastrophic consequences, making the development of effective mitigation techniques essential. In this paper, we introduce MAED (Mathematical Activation Error Detection), an algorithm-level error detection framework that exploits mathematical identities to continuously validate the correctness of non-linear activation function computations at runtime. To the best of our knowledge, this work is the first to integrate algorithm-level error detection techniques to defend against both malicious fault injection attacks and naturally occurring faults in critical DNN components in embedded systems. The evaluation is conducted on three widely adopted activation functions, namely ReLu, sigmoid, and tanh which serve as fundamental building blocks for introducing non-linearity in DNNs and can lead to mispredictions when subjected to natural faults or fault attacks. We assessed the proposed error detection scheme via fault model simulation, achieving close to 100% error detection while mitigating existing fault attacks on DNN inference. Additionally, the overhead introduced by integrating the proposed scheme with the baseline implementation (i.e., without error detection) is validated through implementations on an AMD/Xilinx Artix-7 FPGA and an ATmega328P microcontroller, as well as through integration with TensorFlow. On the microcontroller, the proposed error detection incurs less than 1% clock cycle overhead, while on the FPGA it requires nearly zero additional area, at the cost of approximately a 20% increase in latency for sigmoid and tanh.

cs.CR↗

arXiv2025

An Interactive Framework for Implementing Privacy-Preserving Federated Learning: Experiments on Large Language Models

Federated learning (FL) enhances privacy by keeping user data on local devices. However, emerging attacks have demonstrated that the updates shared by users during training can reveal significant information about their data. This has greatly thwart the adoption of FL methods for training robust AI models in sensitive applications. Differential Privacy (DP) is considered the gold standard for safeguarding user data. However, DP guarantees are highly conservative, providing worst-case privacy guarantees. This can result in overestimating privacy needs, which may compromise the model's accuracy. Additionally, interpretations of these privacy guarantees have proven to be challenging in different contexts. This is further exacerbated when other factors, such as the number of training iterations, data distribution, and specific application requirements, can add further complexity to this problem. In this work, we proposed a framework that integrates a human entity as a privacy practitioner to determine an optimal trade-off between the model's privacy and utility. Our framework is the first to address the variable memory requirement of existing DP methods in FL settings, where resource-limited devices (e.g., cell phones) can participate. To support such settings, we adopt a recent DP method with fixed memory usage to ensure scalable private FL. We evaluated our proposed framework by fine-tuning a BERT-based LLM model using the GLUE dataset (a common approach in literature), leveraging the new accountant, and employing diverse data partitioning strategies to mimic real-world conditions. As a result, we achieved stable memory usage, with an average accuracy reduction of 1.33% for $ε= 10$ and 1.9% for $ε= 6$, when compared to the state-of-the-art DP accountant which does not support fixed memory usage.

cs.LG↗

arXiv2024

Efficient Algorithm Level Error Detection for Number-Theoretic Transform used for Kyber Assessed on FPGAs and ARM

Polynomial multiplication stands out as a highly demanding arithmetic process in the development of post-quantum cryptosystems. The importance of the number-theoretic transform (NTT) extends beyond post-quantum cryptosystems, proving valuable in enhancing existing security protocols such as digital signature schemes and hash functions. CRYSTALS-KYBER stands out as the sole public key encryption (PKE) algorithm chosen by the National Institute of Standards and Technology (NIST) in its third round selection, making it highly regarded as a leading post-quantum cryptography (PQC) solution. Due to the potential for errors to significantly disrupt the operation of secure, cryptographically-protected systems, compromising data integrity, and safeguarding against side-channel attacks initiated through faults it is essential to incorporate mitigating error detection schemes. This paper introduces algorithm level fault detection schemes in the NTT multiplication using Negative Wrapped Convolution and the NTT tailored for Kyber Round 3, representing a significant enhancement compared to previous research. We evaluate this through the simulation of a fault model, ensuring that the conducted assessments accurately mirror the obtained results. Consequently, we attain a notably comprehensive coverage of errors. Furthermore, we assess the performance of our efficient error detection scheme for Negative Wrapped Convolution on FPGAs to showcase its implementation and resource requirements. Through implementation of our error detection approach on Xilinx/AMD Zynq Ultrascale+ and Artix-7, we achieve a comparable throughput with just a 9% increase in area and 13% increase in latency compared to the original hardware implementations. Finally, we attained an error detection ratio of nearly 100% for the NTT operation in Kyber Round 3, with a clock cycle overhead of 16% on the Cortex-A72 processor.

cs.CR↗

arXiv2024

Efficient Fault Detection Architectures for Modular Exponentiation Targeting Cryptographic Applications Benchmarked on FPGAs

Whether stemming from malicious intent or natural occurrences, faults and errors can significantly undermine the reliability of any architecture. In response to this challenge, fault detection assumes a pivotal role in ensuring the secure deployment of cryptosystems. Even when a cryptosystem boasts mathematical security, its practical implementation may remain susceptible to exploitation through side-channel attacks. In this paper, we propose a lightweight fault detection architecture tailored for modular exponentiation, a building block of numerous cryptographic applications spanning from classical cryptography to post quantum cryptography. Based on our simulation and implementation results on ARM Cortex-A72 processor, and AMD/Xilinx Zynq Ultrascale+, and Artix-7 FPGAs, our approach achieves an error detection rate close to 100%, all while introducing a modest computational overhead of approximately 7% and area overhead of less than 1% compared to the unprotected architecture. To the best of our knowledge, such an approach benchmarked on ARM processor and FPGA has not been proposed and assessed to date.

cs.CR↗

arXiv2023

Envisioning the Future of Cyber Security in Post-Quantum Era: A Survey on PQ Standardization, Applications, Challenges and Opportunities

The rise of quantum computers exposes vulnerabilities in current public key cryptographic protocols, necessitating the development of secure post-quantum (PQ) schemes. Hence, we conduct a comprehensive study on various PQ approaches, covering the constructional design, structural vulnerabilities, and offer security assessments, implementation evaluations, and a particular focus on side-channel attacks. We analyze global standardization processes, evaluate their metrics in relation to real-world applications, and primarily focus on standardized PQ schemes, selected additional signature competition candidates, and PQ-secure cutting-edge schemes beyond standardization. Finally, we present visions and potential future directions for a seamless transition to the PQ era.

cs.CR↗