Safe Collaborative Filtering
This work addresses algorithmic fairness and user retention in personalized recommender systems by focusing on tail performance, though it is incremental as it extends the scalable iALS method.
The paper tackles the problem of improving recommendation quality for less-satisfied users in collaborative filtering by minimizing conditional value at risk (CVaR) to enhance tail performance, demonstrating excellent tail results on real-world datasets while maintaining competitive computational efficiency.
Excellent tail performance is crucial for modern machine learning tasks, such as algorithmic fairness, class imbalance, and risk-sensitive decision making, as it ensures the effective handling of challenging samples within a dataset. Tail performance is also a vital determinant of success for personalized recommender systems to reduce the risk of losing users with low satisfaction. This study introduces a "safe" collaborative filtering method that prioritizes recommendation quality for less-satisfied users rather than focusing on the average performance. Our approach minimizes the conditional value at risk (CVaR), which represents the average risk over the tails of users' loss. To overcome computational challenges for web-scale recommender systems, we develop a robust yet practical algorithm that extends the most scalable method, implicit alternating least squares (iALS). Empirical evaluation on real-world datasets demonstrates the excellent tail performance of our approach while maintaining competitive computational efficiency.