ML challenge inspired from the aggregate reporting API proposal

[BasileLeparmentier]

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 dataset and some prize money and let researchers and data scientists from around the world compete on how to learn performing bidding models from differential private reports. Link to the challenge here.
​
We will be happy to work with the Chrome team to set the appropriate parameter to the differential privacy function, to be as close as possible to what real-life operations could look like (e.g. an epsilon level would be beneficial).
​
We hope to kickstart the challenge in early May, so if you are interested in solving the “optimisation” use case using the aggregate reporting API, please do participate!
​
Best,
​
Basile on behalf of the Criteo team

[BasileLeparmentier]

Dear @csharrison ,

For the challenge could you please provide us for an order of magnitude of the epsilon for us to work with? If you have an idea on the noise distribution too (Gaussian? Laplacian?) it would be great!

We understand very well that this number might change (and we hope the challenge will help us inform the choice of epsilon!) but it would really help to have an order of magnitude to set up the challenge.

Many thanks!

[michaelkleber]

(Hi @BasileLeparmentier, I think this issue ought to be on the https://github.com/WICG/conversion-measurement-api repo instead of this one, though Charlie will probably see it here too)

[BasileLeparmentier]

Thank you!