Weakly supervised deep metric learning on discrete metric spaces for privacy-preserved clustering
作者：
Highlights：
• Proposed an encoding-based workflow of data clustering that preserves data privacy.
• Proposed solution is suitable for deployment in a distributed computing environment.
• A weakly supervised approach is used to learn a parameterized similarity function.
• Effective reconstruction of data is achived with additional statistical information.
• Experiments on image and text datasets shows the efficacy of the proposed method.
摘要
•Proposed an encoding-based workflow of data clustering that preserves data privacy.•Proposed solution is suitable for deployment in a distributed computing environment.•A weakly supervised approach is used to learn a parameterized similarity function.•Effective reconstruction of data is achived with additional statistical information.•Experiments on image and text datasets shows the efficacy of the proposed method.
论文关键词：Privacy preservation,K-means clustering,Deep metric learning,Triplet network
论文评审过程：Received 2 January 2022, Revised 5 August 2022, Accepted 28 September 2022, Available online 19 October 2022, Version of Record 19 October 2022.
论文官网地址：https://doi.org/10.1016/j.ipm.2022.103109
