uCTRL: Unbiased Contrastive Representation Learning via Alignment and Uniformity for Collaborative Filtering
This addresses popularity bias in recommender systems for users, offering a novel approach that improves over existing unbiased methods.
The paper tackles the problem of popularity bias in collaborative filtering models by proposing uCTRL, an unbiased contrastive representation learning method that optimizes alignment and uniformity functions, resulting in up to 12.22% and 16.33% gains in Recall@20 and NDCG@20 on benchmark datasets.
Because implicit user feedback for the collaborative filtering (CF) models is biased toward popular items, CF models tend to yield recommendation lists with popularity bias. Previous studies have utilized inverse propensity weighting (IPW) or causal inference to mitigate this problem. However, they solely employ pointwise or pairwise loss functions and neglect to adopt a contrastive loss function for learning meaningful user and item representations. In this paper, we propose Unbiased ConTrastive Representation Learning (uCTRL), optimizing alignment and uniformity functions derived from the InfoNCE loss function for CF models. Specifically, we formulate an unbiased alignment function used in uCTRL. We also devise a novel IPW estimation method that removes the bias of both users and items. Despite its simplicity, uCTRL equipped with existing CF models consistently outperforms state-of-the-art unbiased recommender models, up to 12.22% for Recall@20 and 16.33% for NDCG@20 gains, on four benchmark datasets.