DRanker: A Transformer-Based Multi-Task Ranking Model for the Disney+ Homepage
DRanker is a transformer-based ranking model designed to personalize content ordering for a streaming service homepage. It uses multi-task learning to predict both user engagement probability and expected engagement length simultaneously. Key improvements included using first interactions only, stratified sampling, and a causal debiasing tower to mitigate popularity bias. The model achieved significant lifts in key engagement metrics.