• Lakshmi Manoharan | Senior Machine Learning Engineer, Ads Vertical Modeling; Karthik Jayasurya | Staff Machine Learning Engineer, Ads Signals ; Ziwei Guo | Senior Machine Learning Engineer, Ads Vertical Modeling; Joy Xin | Machine Learning Engineer II, Ads Vertical Modeling; Alina Liviniuk | Machine Learning Engineer II, Ads Vertical Modeling Context At Pinterest, ads are more than just advertisements; they are a vital part of the content ecosystem, designed to inspire users and connect them with products and ideas they love. • Our goal is to surface the right ads at the right time, ensuring they seamlessly integrate into a user’s shopping journey and provide genuine value. • To achieve this, understanding user behavior is paramount. • Delivering highly relevant ads in a dynamic environment like Pinterest presents unique challenges. • Users’ interests and shopping intents evolve rapidly, making it crucial for our ad systems to adapt and anticipate their needs. • Traditional ad targeting methods often rely on broad demographic data or static interest categories, which can fall short in capturing the nuanced and evolving nature of user behavior.

Article Summaries:

  • Pinterest’s Ads team has introduced a transformer‑based behavioral sequence model to improve ad candidate generation. The two‑tower architecture encodes users’ off‑site event histories with a bidirectional transformer, while an MLP processes advertiser representations. Training uses in‑batch negatives and sampled softmax with log‑Q bias correction to mitigate popularity bias. The model predicts which advertisers a user is likely to interact with next, based on future conversion events (checkout, add‑to‑cart, signup). Offline evaluation focuses on Recall@K, measuring how well the system retrieves relevant advertisers from a large candidate set. This approach aims to deliver more personalized, timely ads on Pinterest.

Sources: