-
Hi everyone,
We are delighted to announce that we will be organising a challenge with adKDD inspired by the aggregate measurement API, tackling the optimisation use case.
Criteo will provide a…
-
By Abadi et al. theorem 1 (pg. 4) we can choose constants c1 and c2 such that sigma will guarantee differential privacy
-
https://doi.org/10.1101/159756 (http://www.biorxiv.org/content/early/2017/07/05/159756.1)
> Though it is widely recognized that data sharing enables faster scientific progress, the sensible need to…
-
### Description
The \epsilon and/or \varepsilon characters are poorly rendered, almost not visible. This is a big issue for papers on differential privacy, where epsilon is the key parameter.
### (O…
-
In the last PAT-CG meeting, I mentioned the possibility of a different type of IPA query, where instead of passing in just one "breakdown_key" per source-event, a report collector could instead specif…
-
pp.4-5, again, possibly a sweeping statement about anonymisation. Yes, anonymisation can never be perfect or undoable. But be careful not to imply that it is a useless tool. Differential privacy is pr…
-
In [PrivVG](https://github.com/privvg) project, document frequencies will be necessary in order to satisfy the differential privacy definition (a single new document might inflate counts a lot if it v…
-
I'm a little bit confused by subsections **Secure Deep Learning** and **General Research** - I think it would make sense to split the awesome-list subsections into current research directions. I propo…
-
Write a paper or technical report that compares differentially-private synthesis methods for their utility, as the privacy parameter (epsilon) varies.
- identify suitable synthesis methods (based o…
ots22 updated
4 years ago
-
Determine whether it is possible to encrypt the audio, and the learned audio/model using an encryption to which nobody has the key, to learn on homogenously encrypted data without sharing audio data w…