TY - GEN
T1 - PPSL
T2 - 4th International Conference on Data Intelligence and Security, ICDIS 2022
AU - Alnasser, Walaa
AU - Beigi, Ghazaleh
AU - Mosallanezhad, Ahmadreza
AU - Liu, Huan
N1 - Funding Information: This work is supported by the Saudi Arabian Cultural Mission (SACM) in the United States; National Science Foundation (NSF) grants #2114789; and ONR N00014-21-1-4002. The views, opinions and/or findings expressed are the authors’ and should not be interpreted as representing the official views or policies of the U.S. Government or ONR. Publisher Copyright: © 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - With the blooming of machine learning, Distributed Collaborative Machine Learning (DCML) approaches have been used in various applications to scale up the training process. However, they may have privacy issues that need to be addressed. Split learning is one of the latest DCML approaches that enable the training of the machine learning models without sharing the raw data. In this paper, we study the novel problem of building a privacy-preserving text classifier in the split learning setting. This task, however, is challenging due to the need to maintain the utility of the text data for downstream tasks while protecting privacy by preventing the leakage of private attributes. We address this dilemma of privacy and utility in this work. We propose a text classification for split learning, PPSL, which utilizes the adversarial learning to minimize the private attribute leakage. We study the impacts of increasing the training time and the number of hidden layers on the privacy of split learning. Our experimental results demonstrate the effectiveness of the proposed framework that retains the sentiment meaning and preserves the private attributes while minimizing the leakage.
AB - With the blooming of machine learning, Distributed Collaborative Machine Learning (DCML) approaches have been used in various applications to scale up the training process. However, they may have privacy issues that need to be addressed. Split learning is one of the latest DCML approaches that enable the training of the machine learning models without sharing the raw data. In this paper, we study the novel problem of building a privacy-preserving text classifier in the split learning setting. This task, however, is challenging due to the need to maintain the utility of the text data for downstream tasks while protecting privacy by preventing the leakage of private attributes. We address this dilemma of privacy and utility in this work. We propose a text classification for split learning, PPSL, which utilizes the adversarial learning to minimize the private attribute leakage. We study the impacts of increasing the training time and the number of hidden layers on the privacy of split learning. Our experimental results demonstrate the effectiveness of the proposed framework that retains the sentiment meaning and preserves the private attributes while minimizing the leakage.
KW - adversarial learning
KW - privacy
KW - split learning
KW - text
UR - http://www.scopus.com/inward/record.url?scp=85146494189&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85146494189&partnerID=8YFLogxK
U2 - 10.1109/ICDIS55630.2022.00032
DO - 10.1109/ICDIS55630.2022.00032
M3 - Conference contribution
T3 - Proceedings - 2022 4th International Conference on Data Intelligence and Security, ICDIS 2022
SP - 160
EP - 167
BT - Proceedings - 2022 4th International Conference on Data Intelligence and Security, ICDIS 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 24 August 2022 through 26 August 2022
ER -