WWW 2022 推荐系统和广告相关论文整理分类

关注我们，一起学习~

WWW 2022接收论文已经发布。这次共收到了1822篇论文，接收323篇，录用率为17.7%。完整清单见下面的链接

https://www2022.thewebconf.org/accepted-papers/

本文对广告和推荐相关的文章进行了整理和分类，广告的分在了一个大类，推荐相关的分为了冷启动，会话推荐，序列推荐，CTR预测相关，纠偏相关，可解释性，采样方法等领域，根据使用的技术方法分为了图学习，因果关系，强化学习，多任务学习等希望对大家有所帮助。

广告

Equilibria in Auctions with Ad Types【广告类型的拍卖均衡】

On Designing a Two-stage Auction for Online Advertising【论网络广告的两阶段拍卖设计】

Auction design in an autobidding setting: Randomization improves efficiency beyond VCG【自动出价设置中的拍卖设计：随机化提高了 VCG 之外的效率】

Auctions Between Regret Minimizing Agents【遗憾最小化代理之间的拍卖】

Beyond Customer Lifetime Valuation: Measuring the Value of Acquisition and Retention for Subscription Services【超越客户终身估值：衡量订阅服务的获取和保留价值】

Calibrated Click-Through Auctions【校准的点击式拍卖】

Cross DQN: Cross Deep Q Network for Ads Allocation in Feed【Cross DQN：用于 Feed 中广告分配的 Cross Deep Q Network】

Nash Convergence of Mean-Based Learning Algorithms in First Price Auctions【首次价格拍卖中基于均值的学习算法的纳什收敛】

Price Manipulability in First-Price Auctions【首价拍卖中的价格操纵性】

The Parity Ray Regularizer for Pacing in Auction Markets【用于拍卖市场节奏的 Parity Ray 正则化器】

推荐

冷启动

Alleviating Cold-start Problem in CTR Prediction with A Variational Embedding Learning Framework【使用变分嵌入学习框架缓解 CTR 预测中的冷启动问题】

PNMTA: A Pretrained Network Modulation and Task Adaptation Approach for User Cold-Start Recommendation【PMNTA：一种用于用户冷启动推荐的预训练网络调制和任务适应方法】

KoMen: Domain Knowledge Guided Interaction Recommendation for Emerging Scenarios【KoMen：新兴场景的领域知识引导交互推荐】

点击CTR

Alleviating Cold-start Problem in CTR Prediction with A Variational Embedding Learning Framework【使用变分嵌入学习框架缓解 CTR 预测中的冷启动问题】

Deep Interest Highlight Network for Click-Through Rate Prediction in Trigger-Induced Recommendation【触发诱导推荐中点击率预测的深度兴趣突出网络】

ParClick: A Scalable Algorithm for EM-based Click Models【ParClick：基于 EM 的 Click 模型的可扩展算法】

CBR: Context Bias aware Recommendation for Debiasing User Modeling and Click Prediction【CBR：用于消除用户建模和点击预测的上下文偏差感知建议】

Rating Distribution Calibration for Selection Bias Mitigation in Recommendations【推荐中减少选择偏差的评级分布校准】

MBCT: Tree-Based Feature-Aware Binning for Individual Uncertainty Calibration【MBCT：用于个体不确定性校准的基于树的特征感知分箱】

session会话推荐

Generative Session-based Recommendation【基于生成会话的推荐】

GSL4Rec: Session-based Recommendations with Collective Graph Structure Learning and Next Interaction Prediction【GSL4Rec：具有集体图结构学习和下一次交互预测的基于会话的推荐】

sequential序列推荐

Efficient Online Learning to Rank for Sequential Music Recommendation【高效的在线学习对顺序音乐推荐进行排名】

Filter-enhanced MLP is All You Need for Sequential Recommendation【过滤器增强的 MLP 是进行顺序推荐所需的全部】

Intent Contrastive Learning for Sequential Recommendation【序列推荐的意图对比学习】

Learn from Past, Evolve for Future: Search-based Time-aware Recommendation with Sequential Behavior Data【从过去学习，为未来发展：基于搜索的时间感知推荐与顺序行为数据】

Sequential Recommendation via Stochastic Self-Attention【通过随机自注意力的顺序推荐】

Sequential Recommendation with Decomposed Item Feature Routing【具有分解项目特征路由的顺序推荐】

Towards Automatic Discovering of Deep Hybrid Network Architecture for Sequential Recommendation【面向顺序推荐的深度混合网络架构的自动发现】

Unbiased Sequential Recommendation with Latent Confounders【具有潜在混杂因素的无偏顺序推荐】

Disentangling Long and Short-Term Interests for Recommendation【解耦推荐的长期和短期利益】

跨域

Collaborative Filtering with Attribution Alignment for Review-based Non-overlapped Cross Domain Recommendation【用于基于评论的非重叠跨域推荐的具有属性对齐的协同过滤】

Differential Private Knowledge Transfer for Privacy-Preserving Cross-Domain Recommendation【隐私保护跨域推荐的差分私有知识迁移】

Learning Robust Recommenders through Cross-Model Agreement【通过跨模型协议学习强大的推荐器】

纠偏

Cross Pairwise Ranking for Unbiased Item Recommendation【无偏项目推荐的交叉成对排名】

CBR: Context Bias aware Recommendation for Debiasing User Modeling and Click Prediction【CBR：用于消除用户建模和点击预测的上下文偏差感知建议】

Unbiased Sequential Recommendation with Latent Confounders【具有潜在混杂因素的无偏顺序推荐】

Asymptotically Unbiased Estimation for Delayed Feedback Modeling via Label Correction【通过标签校正的延迟反馈建模的渐近无偏估计】

UKD: Debiasing Conversion Rate Estimation via Uncertainty-regularized Knowledge Distillation【UKD：通过不确定性正则化知识蒸馏的去偏转换率估计】

个性化

Regulatory Instruments for Fair Personalized Pricing【公平个性化定价的监管工具】

Discovering Personalized Semantics for Soft Attributes in Recommender Systems Using Concept Activation Vectors【使用概念激活向量发现推荐系统中软属性的个性化语义】

Improving Personalized Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks【通过调整辅助任务的梯度幅度来改进个性化推荐】

An Empirical Investigation of Personalization Factors on TikTok【TikTok个性化因素的实证研究】

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于图神经网络的个性化可视化推荐】

因果关系

A Model-Agnostic Causal Learning Framework for Recommendation using Search Data【使用搜索数据进行推荐的与模型无关的因果学习框架】

Causal Preference Learning for Out-of-Distribution Recommendation【分布外推荐的因果偏好学习】

Unbiased Sequential Recommendation with Latent Confounders【具有潜在混杂因素的无偏顺序推荐】

可解释

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于图神经网络的个性化可视化推荐】

Path Language Modeling over Knowledge Graphs for Explainable Recommendation【可解释推荐的知识图路径语言建模】

Accurate and Explainable Recommendation via Review Rationalization【通过审查合理化提供准确且可解释的建议】

AmpSum: Adaptive Multiple-Product Summarization towards Improving Recommendation Explainability【AmpSum：提高推荐可解释性的自适应多产品总结】

Comparative Explanations of Recommendations【推荐系统的比较解释】

Neuro-Symbolic Interpretable Collaborative Filtering for Attribute-based Recommendation【基于属性推荐的神经符号可解释协同过滤】

公平性

FairGAN: GANs-based Fairness-aware Learning for Recommendations with Implicit Feedback【FairGAN：基于 GAN 的公平感知学习，用于具有隐式反馈的推荐】

Regulatory Instruments for Fair Personalized Pricing【公平个性化定价的监管工具】

Differential Private Knowledge Transfer for Privacy-Preserving Cross-Domain Recommendation【隐私保护跨域推荐的差分私有知识迁移

多任务

A Multi-task Learning Framework for Product Ranking with BERT【使用 BERT 进行产品排名的多任务学习框架】

A Contrastive Sharing Model for Multi-Task Recommendation【多任务推荐的对比共享模型】

对比学习

Intent Contrastive Learning for Sequential Recommendation【序列推荐的意图对比学习】

A Contrastive Sharing Model for Multi-Task Recommendation【多任务推荐的对比共享模型】

Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning【使用邻域丰富的对比学习改进图协同过滤】

图学习

Hypercomplex Graph Collaborative Filtering【超复杂图协同过滤】

Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning【使用邻域丰富的对比学习改进图协同过滤】

STAM: A Spatiotemporal Aggregation Method for Graph Neural Network-based Recommendation【STAM：一种基于图神经网络推荐的时空聚合方法】

FIRE: Fast Incremental Recommendation with Graph Signal Processing【FIRE：使用图信号处理的快速增量推荐】

Graph Neural Transport Networks with Non-local Attentions for Recommender Systems【用于推荐系统的具有非局部注意力的图神经传输网络】

Multi-level Recommendation Reasoning over Knowledge Graphs with Reinforcement Learning【基于强化学习的知识图的多级推荐推理】

Revisiting Graph Neural Network based Social Recommendation【重新审视基于图神经网络的社交推荐】

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于图神经网络的个性化可视化推荐】

Path Language Modeling over Knowledge Graphs for Explainable Recommendation【可解释推荐的知识图路径语言建模】

GSL4Rec: Session-based Recommendations with Collective Graph Structure Learning and Next Interaction Prediction【GSL4Rec：具有集体图结构学习和下一次交互预测的基于会话的推荐】

Optimizing Rankings for Recommendation in Matching Markets【优化匹配市场中推荐的排名】Yi Su, Magd Bayoumi and Thorsten Joachims

采样

Learning Recommenders for Implicit Feedback with Importance Resampling【通过重要性重采样学习隐式反馈的推荐器】

A Gain-Tuning Dynamic Negative Sampler for Recommendation【用于推荐的增益调整动态负采样器】Qiannan Zhu, Haobo Zhang, Qing He and Zhicheng Dou

新闻

FeedRec: News Feed Recommendation with Various User Feedbacks【FeedRec：具有各种用户反馈的新闻提要推荐】

MINDSim: User Simulator for News Recommenders【MINDSim：新闻推荐者的用户模拟器】

在线学习

Learning Neural Ranking Models Online from Implicit User Feedback【从隐式用户反馈在线学习神经排序模型】

Efficient Online Learning to Rank for Sequential Music Recommendation【高效的在线学习对顺序音乐推荐进行排名】

强化学习

Multiple Choice Questions based Multi-Interest Policy Learning for Conversational Recommendation【基于多项选择题的会话推荐多兴趣策略学习】

Off-policy Learning over Heterogeneous Information for Recommendation【用于推荐的异构信息的异策略学习】

Multi-level Recommendation Reasoning over Knowledge Graphs with Reinforcement Learning【基于强化学习的知识图的多级推荐推理】

其他

Learning Probabilistic Box Embeddings for Effective and Efficient Ranking【学习用于有效和高效排名的概率框嵌入】

Conditional Generation Net for Medication Recommendation【药物推荐的条件生成网络】

Automating Feature Selection in Deep Recommender Systems【深度推荐系统中的自动特征选择】

Choice of Implicit Signal Matters: Accounting for UserAspirations in Podcast Recommendations【隐式信号的选择很重要：考虑播客推荐中的用户愿望】

Deep Unified Representation for Heterogeneous Recommendation【异构推荐的深度统一表示】

Learning to Augment for Casual User Recommendation【学习增强临时用户推荐】

Modality Matches Modality: Pretraining Modality-Disentangled Item Representations for Recommendation【模态匹配模态：预训练模态分离商品表征以进行推荐】

Mutually-Regularized Dual Collaborative Variational Auto-encoder for Recommendation Systems【推荐系统的相互正则化双协同变分自动编码器】

Re4: Learning to Re-contrast, Re-attend, Re-construct for Multi-interest Recommendation【Re4：学习重新对比、重新参与、重新构建多兴趣推荐】

Consensus Learning from Heterogeneous Objectives for One-Class Collaborative Filtering【一类协同过滤的异构目标的共识学习】

Fast Variational AutoEncoder with Inverted Multi-Index for Collaborative Filtering【用于协同过滤的具有倒置多索引的快速变分自动编码器】

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization【HRCF：通过双曲几何正则化增强协同过滤】

MCL: Mixed-Centric Loss for Collaborative Filtering【MCL：用于协同过滤的混合中心损失】

Stochastic-Expert Variational Autoencoder for Collaborative Filtering【用于协同过滤的随机专家变分自动编码器】

Rewiring what-to-watch-next Recommendations to Reduce Radicalization Pathways【重新制定下一步观看的建议以减少激进化途径】

Recommendation Unlearning

What to Watch Next: Two-side Interactive Networksfor Live Broadcast Recommendation【接下来看什么：直播推荐的双边互动网络】

参考