| HAL-Inria | Publications, software ... of Inria's scientists |
Conference papers
Compressive k-Means with Differential Privacy
Abstract : In the compressive learning framework, one harshly compresses a whole training dataset into a single vector of generalized random moments, the sketch, from which a learning task can subsequently be performed. We prove that this loss of information can be leveraged to design a differentially private mechanism, and study empirically the privacy-utility tradeoff for the k-means clustering problem.
Cited literature [11 references]
https://hal.inria.fr/hal-02154820
Contributor : Antoine Chatalic Connect in order to contact the contributor
Submitted on : Thursday, June 13, 2019 - 9:34:59 AM
Last modification on : Friday, April 8, 2022 - 4:04:02 PM
Contributor : Antoine Chatalic Connect in order to contact the contributor
Submitted on : Thursday, June 13, 2019 - 9:34:59 AM
Last modification on : Friday, April 8, 2022 - 4:04:02 PM
File
spars_final.pdf
Files produced by the author(s)