Practical Multi-Task Learning for Rare Conversions in Ad Tech
This work addresses the challenge of improving ad performance metrics for advertisers and platforms, but it is incremental as it builds on existing MTL methods.
The paper tackled the problem of predicting rare conversion events in online advertising by developing a Multi-Task Learning approach that classifies conversions as rare or frequent, resulting in a 0.69% AUC lift offline and a 2% reduction in Cost per Action online.
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).