A Biased Sampling Method for Imbalanced Personalized Ranking

Lu Yu, Shichao Pei, Feng Zhu, Longfei Li, JUN ZHOU, Chuxu Zhang, Xiangliang Zhang

Abstract: Pairwise ranking models have been widely used to address recom- mendation problems. The basic idea is to learn the rank of users’ preferred items through separating items into positive samples if user-item interactions exist, and negative samples otherwise. Due to the limited number of observable interactions, pairwise ranking models face serious class-imbalance issues. Our theoretical analysis shows that current sampling-based methods cause the vertex-level imbalance problem, which makes the norm of learned item em- beddings towards infinite after a certain training iterations, and consequently results in vanishing gradient and affects the model in- ference results. We thus propose an efficient Vital Negative Sampler (VINS) to alleviate the class-imbalance issue for pairwise ranking model, in particular for deep learning models optimized by gradient methods. The core of VINS is a bias sampler with reject probability that will tend to accept a negative candidate with a larger degree weight than the given positive item. Evaluation results on sev- eral real datasets demonstrate that the proposed sampling method speeds up the training procedure 30% to 50% for ranking models ranging from shallow to deep, while maintaining and even improv- ing the quality of ranking results in top-N item recommendation.