重新评估经典推荐算法BPR，发现实现差异导致性能下降50%。
Revisiting BPR: A Replicability Study of a Common Recommender System Baseline
- 系统分析BPR实现细节，揭示开源代码与原始论文的偏差
- 调优后BPR在真实数据集上接近甚至超越顶尖方法性能
- 适合关注推荐系统基线实验可复现性的研究者参考
贝叶斯个性化排序（BPR）是一种基于矩阵分解的协同过滤方法，常被用作推荐系统研究的基准。然而，许多研究忽略了BPR实现中的细微差别，声称其性能低于新提出的方法。本文全面考察了BPR模型的特性，分析其对性能的影响，并调研了公开的BPR实现。结果表明，这些实现与原始论文存在不一致，导致特定实现性能下降高达50%。通过在真实世界数据集上进行大规模实验并采用现代评估设置，我们证明，在超参数正确调优下，BPR在Top-n推荐任务中可达到接近最先进方法的性能，甚至在某些数据集上表现更优。具体而言，在百万歌曲数据集上，经调优的BPR在二值相关性函数下的NDCG@100指标上相比Mult-VAE统计显著提升10%。
原文摘要 · Abstract (English)
Bayesian Personalized Ranking (BPR), a collaborative filtering approach based on matrix factorization, frequently serves as a benchmark for recommender systems research. However, numerous studies often overlook the nuances of BPR implementation, claiming that it performs worse than newly proposed methods across various tasks. In this paper, we thoroughly examine the features of the BPR model, indicating their impact on its performance, and investigate open-source BPR implementations. Our analysis reveals inconsistencies between these implementations and the original BPR paper, leading to a significant decrease in performance of up to 50% for specific implementations. Furthermore, through extensive experiments on real-world datasets under modern evaluation settings, we demonstrate that with proper tuning of its hyperparameters, the BPR model can achieve performance levels close to state-of-the-art methods on the top-n recommendation tasks and even outperform them on specific datasets. Specifically, on the Million Song Dataset, the BPR model with hyperparameters tuning statistically significantly outperforms Mult-VAE by 10% in NDCG@100 with binary relevance function.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。