Learned Ranking Function: From Short-term Behavior Predictions to Long-term User Satisfaction
This work addresses the challenge of improving user retention and engagement in large-scale platforms like YouTube by moving beyond heuristic-based methods.
The paper tackles the problem of optimizing long-term user satisfaction in recommendation systems by introducing the Learned Ranking Function (LRF), which uses short-term behavior predictions to generate slates, and reports deployment on YouTube with live experiments.
We present the Learned Ranking Function (LRF), a system that takes short-term user-item behavior predictions as input and outputs a slate of recommendations that directly optimizes for long-term user satisfaction. Most previous work is based on optimizing the hyperparameters of a heuristic function. We propose to model the problem directly as a slate optimization problem with the objective of maximizing long-term user satisfaction. We also develop a novel constraint optimization algorithm that stabilizes objective trade-offs for multi-objective optimization. We evaluate our approach with live experiments and describe its deployment on YouTube.