Bayesian Negative Sampling for Recommendation

TL;DR

This paper introduces Bayesian Negative Sampling (BNS), a novel method for discriminating true negatives from false negatives in negative sampling, improving recommendation quality with a model-agnostic approach.

Contribution

It develops a Bayesian classifier based on order relation analysis to accurately identify true negatives, enabling more effective negative sampling for recommendation models.

Findings

BNS outperforms existing methods in sampling quality.

BNS improves recommendation accuracy.

The algorithm has linear time complexity.

Abstract

How to sample high quality negative instances from unlabeled data, i.e., negative sampling, is important for training implicit collaborative filtering and contrastive learning models. Although previous studies have proposed some approaches to sample informative instances, few has been done to discriminating false negative from true negative for unbiased negative sampling. On the basis of our order relation analysis of negatives' scores, we first derive the class conditional density of true negatives and that of false negatives. We next design a Bayesian classifier for negative classification, from which we define a model-agnostic posterior probability estimate of an instance being true negative as a quantitative negative signal measure. We also propose a Bayesian optimal sampling rule to sample high-quality negatives. The proposed Bayesian Negative Sampling (BNS) algorithm has a linear time complexity. Experimental studies validate the superiority of BNS over the peers in terms of better sampling quality and better recommendation performance.