Configuring differential privacy policy (optional)

Note

Applies to: AWS Clean Rooms SQL analytics engine

This procedure describes the process of configuring the differential privacy policy in a collaboration by using the Guided flow option in the AWS Clean Rooms console. This is a one-time step for all tables with differential privacy protection.

To configure differential privacy settings (guided flow)

1. Sign in to the AWS Management Console and open the AWS Clean Rooms console with your AWS account (if you haven't yet done so).

2. In the left navigation pane, choose Collaborations.

3. Choose the collaboration.

4. On the Tables tab of the collaboration page, choose Configure differential privacy policy.

5. On the Configure differential privacy policy page, choose values for the following properties:

   • Privacy budget

   • Refresh privacy budget monthly

   • Noise added per query

   You can use the default values or enter custom values that support your specific use case. After choosing values for Privacy budget and Noise added per query, you can preview the resulting utility in terms of the number of aggregations that are possible across all queries on your data.

6. Choose Configure.

You’ll see a confirmation message that you’ve successfully configured the differential privacy policy for the collaboration.

Now that you configured differential privacy, you are ready to:

• Query the data tables (as a member who can query)

• Collaborations (if you're the collaboration creator)

Viewing differential privacy usage logs

As a collaboration member who is protecting data with differential privacy, after you have created a collaboration with differential privacy, you can monitor the usage of the privacy budget.

To view how many aggregations were run and how much of the privacy budget was used

1. Sign in to the AWS Management Console and open the AWS Clean Rooms console with your AWS account (if you have not yet done so).

2. In the left navigation pane, choose Collaborations.

3. Choose the collaboration.

4. Choose the Tables tab.

5. Choose View usage logs (blue text).

6. View the usage details, including the privacy budget and how much utility was provided.

Editing a differential privacy policy

At any time after configuring the differential privacy policy, you can update it to better reflect your privacy needs.

To edit the differential privacy policy

1. Sign in to the AWS Management Console and open the AWS Clean Rooms console with your AWS account (if you have not yet done so).

2. In the left navigation pane, choose Collaborations.

3. Choose the collaboration.

4. On the Tables tab of the collaboration page, under Tables associated by you, choose Edit.

5. On the Edit differential privacy page, choose new values for the following properties:

   • Privacy budget – Move the slider bar to either increase or decrease the budget at any point during a collaboration. You can't decrease the budget after the member who can query has started querying your data. If the Privacy budget is increased, AWS Clean Rooms will continue using the existing budget until it is fully consumed before utilizing the newly added privacy budget.

   • Noise added per query – Move the slider bar to either increase or decrease the Noise added per query at any point during a collaboration.

   Note

   You can chose Interactive examples to explore how different values of Privacy budget and Noise added per query affect the number of aggregate functions that you can run.

   You can't change the value of the Privacy budget refresh. To change your selection, you must delete the differential privacy policy and create a new one.

6. Choose Save changes.

You see a confirmation message that you’ve successfully edited the differential privacy policy.

Deleting a differential privacy policy

You can delete the differential privacy policy from the Tables tab of a collaboration.

To delete the differential privacy policy

1. Sign in to the AWS Management Console and open the AWS Clean Rooms console with your AWS account (if you have not yet done so).

2. In the left navigation pane, choose Collaborations.

3. Choose the collaboration.

4. On the Tables tab of the collaboration page, next to Differential privacy policy, select Delete.

5. If you’re certain that you want to delete the differential privacy policy, choose Delete.

After deleting a differential privacy policy, you can't access the privacy budget usage logs from that policy. Tables with differential privacy turned on can't be queried if the differential privacy policy is deleted.

Viewing the calculated differential privacy parameters

For users with expertise in differential privacy, you can view the calculated differential privacy parameters from the Queries tab of a collaboration.

To view the calculated differential privacy parameters

1. Sign in to the AWS Management Console and open the AWS Clean Rooms console with your AWS account (if you have not yet done so).

2. In the left navigation pane, choose Collaborations.

3. Choose the collaboration.

4. On the Queries tab, in the Results section, select View calculated differential privacy parameters.

In the Calculated differential privacy parameters table, you can see sensitivity values of aggregate functions, which is defined as the maximum amount by which the result of a function can change if a single user's records are added, removed, or modified. The list includes the following differential privacy parameters:

• User contribution limit (UCL) is the maximum number of rows contributed by a user in a SQL query. For example, if you want to count the total number of matched impressions in a specified campaign where each user can have multiple impressions, AWS Clean Rooms Differential Privacy needs to bound the number of impressions of a single user in order to ensure that the differential privacy calculation is accurate. In other words, if any user has more impressions than the bound, then AWS Clean Rooms automatically takes a uniform random sample of that user's impressions as per the computed UCL value and exclude the remaining impressions of that user while executing the query. The UCL value equals to 1 if you are counting the number of unique users. This is because adding, removing, or modifying a single user can change the count of distinct users by at most 1.

• Minimum value is the lower bound of an expression used within an aggregate function such as sum(). For example, if the expression is a column known as purchase_value, minimum value is the lower bound of the column.

• Maximum value is the upper bound of an expression used within an aggregate function such as sum(). For example, if the expression is a column known as purchase_value, maximum value is the upper bound of the column.

In the Calculated differential privacy parameters table, you can use these parameters to better understand the total amount of noise in query results. For example, when the configured Noise added per query is 30 users and a COUNT DISTINCT (user_id) query is run, then AWS Clean Rooms Differential Privacy adds random noise that falls between -30 and 30 with high probability because the sensitivity of COUNT DISTINCT is 1. In the case of a COUNT query with the same configuration, AWS Clean Rooms Differential Privacy adds statistical noise that is scaled by the user contribution limit because a single user could contribute multiple rows to the query result. In the case of SUM query like SUM (purchase_value) where all the column values are positive, the total noise is scaled by the user contribution limit times the maximum value. AWS Clean Rooms Differential Privacy automatically computes the sensitivity parameters to perform noise addition at query run-time and depletes the privacy budget. The depletion of privacy budget is required because the sensitivity parameters are data-dependent.