In the field of machine learning security, as one of the attack surfaces especially for edge devices, the application of side-channel analysis such as correlation power/electromagnetic analysis (CPA/CEMA) is expanding. Aiming to evaluate the leakage resistance of neural network (NN) model parameters, i.e. weights and biases, we conducted a feasibility study of CPA/CEMA on floating-point (FP) operations, which are the basic operations of NNs. This paper proposes approaches to recover weights and biases using CPA/CEMA on multiplication and addition operations, respectively. It is essential to take into account the characteristics of the IEEE 754 representation in order to realize the recovery with high precision and efficiency. We show that CPA/CEMA on FP operations requires different approaches than traditional CPA/CEMA on cryptographic implementations such as the AES.
Hanae NOZAKI
National Institute of Advanced Industrial Science and Technology (AIST)
Kazukuni KOBARA
National Institute of Advanced Industrial Science and Technology (AIST)
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hanae NOZAKI, Kazukuni KOBARA, "Power Analysis of Floating-Point Operations for Leakage Resistance Evaluation of Neural Network Model Parameters" in IEICE TRANSACTIONS on Fundamentals,
vol. E107-A, no. 3, pp. 331-343, March 2024, doi: 10.1587/transfun.2023CIP0012.
Abstract: In the field of machine learning security, as one of the attack surfaces especially for edge devices, the application of side-channel analysis such as correlation power/electromagnetic analysis (CPA/CEMA) is expanding. Aiming to evaluate the leakage resistance of neural network (NN) model parameters, i.e. weights and biases, we conducted a feasibility study of CPA/CEMA on floating-point (FP) operations, which are the basic operations of NNs. This paper proposes approaches to recover weights and biases using CPA/CEMA on multiplication and addition operations, respectively. It is essential to take into account the characteristics of the IEEE 754 representation in order to realize the recovery with high precision and efficiency. We show that CPA/CEMA on FP operations requires different approaches than traditional CPA/CEMA on cryptographic implementations such as the AES.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2023CIP0012/_p
Copy
@ARTICLE{e107-a_3_331,
author={Hanae NOZAKI, Kazukuni KOBARA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Power Analysis of Floating-Point Operations for Leakage Resistance Evaluation of Neural Network Model Parameters},
year={2024},
volume={E107-A},
number={3},
pages={331-343},
abstract={In the field of machine learning security, as one of the attack surfaces especially for edge devices, the application of side-channel analysis such as correlation power/electromagnetic analysis (CPA/CEMA) is expanding. Aiming to evaluate the leakage resistance of neural network (NN) model parameters, i.e. weights and biases, we conducted a feasibility study of CPA/CEMA on floating-point (FP) operations, which are the basic operations of NNs. This paper proposes approaches to recover weights and biases using CPA/CEMA on multiplication and addition operations, respectively. It is essential to take into account the characteristics of the IEEE 754 representation in order to realize the recovery with high precision and efficiency. We show that CPA/CEMA on FP operations requires different approaches than traditional CPA/CEMA on cryptographic implementations such as the AES.},
keywords={},
doi={10.1587/transfun.2023CIP0012},
ISSN={1745-1337},
month={March},}
Copy
TY - JOUR
TI - Power Analysis of Floating-Point Operations for Leakage Resistance Evaluation of Neural Network Model Parameters
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 331
EP - 343
AU - Hanae NOZAKI
AU - Kazukuni KOBARA
PY - 2024
DO - 10.1587/transfun.2023CIP0012
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E107-A
IS - 3
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - March 2024
AB - In the field of machine learning security, as one of the attack surfaces especially for edge devices, the application of side-channel analysis such as correlation power/electromagnetic analysis (CPA/CEMA) is expanding. Aiming to evaluate the leakage resistance of neural network (NN) model parameters, i.e. weights and biases, we conducted a feasibility study of CPA/CEMA on floating-point (FP) operations, which are the basic operations of NNs. This paper proposes approaches to recover weights and biases using CPA/CEMA on multiplication and addition operations, respectively. It is essential to take into account the characteristics of the IEEE 754 representation in order to realize the recovery with high precision and efficiency. We show that CPA/CEMA on FP operations requires different approaches than traditional CPA/CEMA on cryptographic implementations such as the AES.
ER -