Practical Multi-Task Learning for Rare Conversions in Ad Tech

Abstract

We present a Multi-Task Learning (MTL) approach for improving predictions for rare (e.g., <1%) conversion events in online advertising. The conversions are classified into "rare" or "frequent" types based on historical statistics. The model learns shared representations across all signals while specializing through separate task towers for each type. The approach was tested and fully deployed to production, demonstrating consistent improvements in both offline (0.69% AUC lift) and online KPI performance metric (2% Cost per Action reduction).

Accepted to RecSys 2025

References in corpus (4)

• DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems
• ESCM${}^2$: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimation
• Predicting Different Types of Conversions with Multi-Task Learning in Online Advertising
• Click-Conversion Multi-Task Model with Position Bias Mitigation for Sponsored Search in eCommerce

Cited by in corpus (1)

• RAMP: Robust Ad Recommendation Under Limited Personalized-Feature Availability via Masking and Alignment Pathways