分离用户多元兴趣,提升推荐系统在分布外场景下的点击率预测能力。
Disentangled Interest Network for Out-of-Distribution CTR Prediction
- 从因果视角分解用户兴趣、曝光与点击机制,构建可分离的兴趣表征。
- 在三个真实数据集上,AUC和GAUC提升超0.02,对数损失降低13.7%以上。
- 适合关注推荐系统鲁棒性与兴趣建模的算法研究者和工业界应用者。
点击率(CTR)预测是在线信息服务中的关键任务,旨在估计用户点击某条内容的概率。现有方法常假设训练与测试数据来自相同分布,但用户兴趣持续演化导致数据分布变化,引发分布外(OOD)问题。此外,用户通常拥有多个兴趣,且部分兴趣变化更快。为此,本文提出解耦点击率预测模型(DiseCTR),引入推荐的因果视角,分离用户兴趣的多方面特征以缓解推荐中的分布外问题。通过用户兴趣、曝光模型与点击模型的因果分解,设计深度学习实现三类因果机制。具体而言,首先使用稀疏注意力兴趣编码器将原始特征映射为用户兴趣,再引入弱监督兴趣解耦模块学习独立的兴趣嵌表示,最后通过注意力聚合器进行融合预测。在三个真实世界数据集上的实验表明,DiseCTR在分布外推荐中达到最优准确率与鲁棒性,相比先进方法,AUC与GAUC显著提升超过0.02,对数损失降低超过13.7%。进一步分析证实,DiseCTR成功实现了用户兴趣的解耦,这是提升CTR预测分布外泛化能力的关键。代码与数据已公开于 https://github.com/DavyMorgan/DiseCTR/。
原文摘要 · Abstract (English)
Click-through rate (CTR) prediction, which estimates the probability of a user clicking on a given item, is a critical task for online information services. Existing approaches often make strong assumptions that training and test data come from the same distribution. However, the data distribution varies since user interests are constantly evolving, resulting in the out-of-distribution (OOD) issue. In addition, users tend to have multiple interests, some of which evolve faster than others. Towards this end, we propose Disentangled Click-Through Rate prediction (DiseCTR), which introduces a causal perspective of recommendation and disentangles multiple aspects of user interests to alleviate the OOD issue in recommendation. We conduct a causal factorization of CTR prediction involving user interest, exposure model, and click model, based on which we develop a deep learning implementation for these three causal mechanisms. Specifically, we first design an interest encoder with sparse attention which maps raw features to user interests, and then introduce a weakly supervised interest disentangler to learn independent interest embeddings, which are further integrated by an attentive interest aggregator for prediction. Experimental results on three real-world datasets show that DiseCTR achieves the best accuracy and robustness in OOD recommendation against state-of-the-art approaches, significantly improving AUC and GAUC by over 0.02 and reducing logloss by over 13.7%. Further analyses demonstrate that DiseCTR successfully disentangles user interests, which is the key to OOD generalization for CTR prediction. We have released the code and data at https://github.com/DavyMorgan/DiseCTR/.
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。