Meta’s Private Lift Measurement uses MPC to allow both Meta and an advertiser to learn insights about how an ad is performing without the need to trust either party with both datasets. MPC limits the information that can be learned by either Meta or the advertiser.
Code
GitHub - facebookresearch/fbpcs: FBPCS (Facebook Private Computation Solutions)
GitHub - facebookresearch/fbpcf: Private computation framework library
Links
Meta: The Value of Secure Multi-Party Computation
Meta: Privacy-Enhancing Technologies and Building for the Future
Meta: What Are Privacy-Enhancing Technologies (PETs) and How Will They Apply to Ads?