Learnable Audio Encryption for Untrusted Outsourcing Machine Learning Services

被引：0
作者：
Chi, Po-Wen [1 ]
Hsiao, Pin-Hsin [1 ]
机构：
[1] Natl Taiwan Normal Univ, Dept Comp Sci & Informat Engn, Taipei, Taiwan
来源：
2019 14TH ASIA JOINT CONFERENCE ON INFORMATION SECURITY (ASIAJCIS 2019) | 2019年
关键词：
Learnable Encryption; Audio Encryption; Machine Learning;
D O I：
10.1109/AsiaJCIS.2019.00029
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
Applying machine learning to problems has become an avoidable trend. With the help of machine learning techniques, people can make predictions more accurately and can get more benefits. However, the machine learning technique relies on training from lots of data. That is, data should be open to the machine learning service provider. Considering the user privacy issue, the release of user data is not acceptable. In this paper, we propose an approach to take care both the machine learning feature and the data privacy. We focus on audio data and propose an audio encryption technique to keep audio data credential. In the meantime, we make the encrypted audio be able to be trained though machine learning service providers.
引用
收藏
页码：151 / 156
页数：6
相关论文
共 25 条
  • [21] Rouhani BD, 2018, DES AUT CON, DOI [10.1145/3195970.3196023, 10.1109/DAC.2018.8465894]
  • [22] Sadkhan Sattar B., 2007, 24th Radio National Science Conference (NRSC 2007), P1, DOI 10.1109/NRSC.2007.371364
  • [23] Tanaka M., 2018, CORRABS180400490
  • [24] Yao A. C., 1986, 27th Annual Symposium on Foundations of Computer Science (Cat. No.86CH2354-9), P162, DOI 10.1109/SFCS.1986.25
  • [25] Zhang YL, 2017, PROCEEDINGS OF THE TWENTY-SIXTH ACM SYMPOSIUM ON OPERATING SYSTEMS PRINCIPLES (SOSP '17), P19, DOI [10.1145/3132747.3132768, 10.1109/SP.2017.12]
123