Privacy-Preserving Techniques and Clean Room Results

LiveRamp Clean Room applies a customer-specified k-min threshold (referred to as a "Crowd Size" data control parameter) to every user metric that is reported in all queries executed within clean rooms.

For example, consider the following query. If the crowd size (COUNT) were 100, then the last row in the result set will not be displayed to the user because it does not meet the minimum threshold.

SELECT usm.audience_segment, COUNT(*) AS impressions
FROM adlog_impressions ali, user_segment_map usm
WHERE ali.user_id = usm.user_id
GROUP BY 1
ORDER BY 2 DESC;

In addition to the k-min anonymity enforcement, noise injection (referred to as a "Data Decibel" data control parameter) lets you apply a randomized metric data perturbation technique by adding Laplacian noise to alter the metrics shown to the user. This prevents leakage of sensitive information if a user tries to run queries that would return similar results with a known difference that could be exploited. You can specify the amount of noise using the Data Decibel parameter in the Create New Clean Room wizard.

The noise injection is based on random noise drawn from a Laplace distribution. The following perturbed function satisfies Epsilon (ἐ) differential privacy:

ML(x,f(.),ἐ) = f(x) + Lap(0,Δf/ἐ)

Where Δf/ἐ is the scale, 0 is the center of the distribution, and the value of Δf (also known as sensitivity) is 1 in the case of count metrics. Thus, the scaling factor for noise level is the Epsilon value within the formula.

This methodology is implemented as a Python user-defined function (UDF) appended to clean room queries. Question templates allow clean room owners to specify which metrics should be treated with random noise.

The Epsilon value is adjusted based on where the Data Decibel slider is set. A higher data decibel yields an Epsilon value closer to 0 (more noise) and a lower decibel yields an Epsilon value closer to 1 (less noise).

The following example scenario shows a base value before noise injection of 14,237. Its decibel values change depending on the amount of noise applied using the slider: Low (1), Medium (50), and High (99). Low and medium noise do not result in much variation, whereas high noise significantly changes the values.

When applying k-min anonymity enforcement and noise injection, consider the following: