In this paper, we propose a selective membership inference attack method that determines whether certain data corresponding to a specific class are being used as training data for a machine learning model or not. By using the proposed method, membership or non-membership can be inferred by generating a decision model from the prediction of the inference models and training the confidence values for the data corresponding to the selected class. We used MNIST as an experimental dataset and Tensorflow as a machine learning library. Experimental results show that the proposed method has a 92.4% success rate with 5 inference models for data corresponding to a specific class.
Hyun KWON
Korea Military Academy
Yongchul KIM
Korea Military Academy
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
Hyun KWON, Yongchul KIM, "Toward Selective Membership Inference Attack against Deep Learning Model" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 11, pp. 1911-1915, November 2022, doi: 10.1587/transinf.2022NGL0001.
Abstract: In this paper, we propose a selective membership inference attack method that determines whether certain data corresponding to a specific class are being used as training data for a machine learning model or not. By using the proposed method, membership or non-membership can be inferred by generating a decision model from the prediction of the inference models and training the confidence values for the data corresponding to the selected class. We used MNIST as an experimental dataset and Tensorflow as a machine learning library. Experimental results show that the proposed method has a 92.4% success rate with 5 inference models for data corresponding to a specific class.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022NGL0001/_p
Copy
@ARTICLE{e105-d_11_1911,
author={Hyun KWON, Yongchul KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Toward Selective Membership Inference Attack against Deep Learning Model},
year={2022},
volume={E105-D},
number={11},
pages={1911-1915},
abstract={In this paper, we propose a selective membership inference attack method that determines whether certain data corresponding to a specific class are being used as training data for a machine learning model or not. By using the proposed method, membership or non-membership can be inferred by generating a decision model from the prediction of the inference models and training the confidence values for the data corresponding to the selected class. We used MNIST as an experimental dataset and Tensorflow as a machine learning library. Experimental results show that the proposed method has a 92.4% success rate with 5 inference models for data corresponding to a specific class.},
keywords={},
doi={10.1587/transinf.2022NGL0001},
ISSN={1745-1361},
month={November},}
Copy
TY - JOUR
TI - Toward Selective Membership Inference Attack against Deep Learning Model
T2 - IEICE TRANSACTIONS on Information
SP - 1911
EP - 1915
AU - Hyun KWON
AU - Yongchul KIM
PY - 2022
DO - 10.1587/transinf.2022NGL0001
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2022
AB - In this paper, we propose a selective membership inference attack method that determines whether certain data corresponding to a specific class are being used as training data for a machine learning model or not. By using the proposed method, membership or non-membership can be inferred by generating a decision model from the prediction of the inference models and training the confidence values for the data corresponding to the selected class. We used MNIST as an experimental dataset and Tensorflow as a machine learning library. Experimental results show that the proposed method has a 92.4% success rate with 5 inference models for data corresponding to a specific class.
ER -