← Back to the feed
Chips & computingResearch findings

CriteoPrivateAds Dataset: A Foundation for Designing Privacy-Respecting Advertising Systems

Original published 2025.02.18 03:24 KSTAdded to Blackdeck 2026.10.24 17:19 KST

What has changed?

The largest privacy-constrained advertising bidding dataset is released, enabling evaluation of privacy-preserving advertising frameworks.

Key points

  1. CriteoPrivateAds is the largest dataset designed for privacy-preserving advertising bidding under constraints such as differential privacy.
  2. It aligns with major browser privacy proposals like Chrome Privacy Sandbox, including user-level aggregation and delayed reporting.
  3. Anonymized from Criteo's production logs, it enables offline evaluation of bidding models under real-world privacy constraints.

Connect it to your work

CriteoPrivateAds is built to align with major browser privacy proposals like Chrome Privacy Sandbox, including delayed reports and user-level differential隐私.

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivCriteoPrivateAds: A Real-World Bidding Dataset to Design Private Advertising Systems