Topics in differential privacy: optimal noise and record perturbation baseddata sets
Duration: 1 hour 6 mins
Share this media item:
Embed this media item:
Embed this media item:
About this item
| Description: |
Soria-Comas, J (Universitat Rovira i Virgili)
Friday 2nd December 2016 - 14:00 to 15:00 |
|---|
| Created: | 2017-01-03 16:43 |
|---|---|
| Collection: | Data Linkage and Anonymisation |
| Publisher: | Isaac Newton Institute |
| Copyright: | Soria-Comas, J |
| Language: | eng (English) |
| Abstract: | We explore two different aspects of differential privacy. First we explore the optimality of noise distributions in noise addition. In particular, we show that the Laplace distribution is nearly optimal in the univariate case, but not in the multivariate case. Optimal distributions are described. Then we explore the generation of differentially private data sets via perturbative masking of the original records. This approach is
remarkably more efficient than histogram-based approaches but a naive application of it may completely damage the data utility. In particular, we analyze the use of microaggregation to reduce the sensitivity and, thus, the amount of noise required to attain differential privacy. |
|---|---|
Available Formats
| Format | Quality | Bitrate | Size | |||
|---|---|---|---|---|---|---|
| MPEG-4 Video | 640x360 | 1.93 Mbits/sec | 958.30 MB | View | Download | |
| WebM | 640x360 | 647.24 kbits/sec | 317.62 MB | View | Download | |
| iPod Video | 480x270 | 493.72 kbits/sec | 238.66 MB | View | Download | |
| MP3 | 44100 Hz | 249.98 kbits/sec | 122.67 MB | Listen | Download | |
| Auto * | (Allows browser to choose a format it supports) | |||||