關(guān)注我們，一起學(xué)習(xí)~

WWW 2022接收論文已經(jīng)發(fā)布。這次共收到了1822篇論文，接收323篇，錄用率為17.7%。完整清單見下面的鏈接

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

本文對廣告和推薦相關(guān)的文章進(jìn)行了整理和分類，廣告的分在了一個(gè)大類，推薦相關(guān)的分為了冷啟動(dòng)，會話推薦，序列推薦，CTR預(yù)測相關(guān)，糾偏相關(guān)，可解釋性，采樣方法等領(lǐng)域，根據(jù)使用的技術(shù)方法分為了圖學(xué)習(xí)，因果關(guān)系，強(qiáng)化學(xué)習(xí)，多任務(wù)學(xué)習(xí)等希望對大家有所幫助。

廣告

Equilibria in Auctions with Ad Types【廣告類型的拍賣均衡】

On Designing a Two-stage Auction for Online Advertising【論網(wǎng)絡(luò)廣告的兩階段拍賣設(shè)計(jì)】

Auction design in an autobidding setting: Randomization improves efficiency beyond VCG【自動(dòng)出價(jià)設(shè)置中的拍賣設(shè)計(jì)：隨機(jī)化提高了 VCG 之外的效率】

Auctions Between Regret Minimizing Agents【遺憾最小化代理之間的拍賣】

Beyond Customer Lifetime Valuation: Measuring the Value of Acquisition and Retention for Subscription Services【超越客戶終身估值：衡量訂閱服務(wù)的獲取和保留價(jià)值】

Calibrated Click-Through Auctions【校準(zhǔn)的點(diǎn)擊式拍賣】

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

Nash Convergence of Mean-Based Learning Algorithms in First Price Auctions【首次價(jià)格拍賣中基于均值的學(xué)習(xí)算法的納什收斂】

Price Manipulability in First-Price Auctions【首價(jià)拍賣中的價(jià)格操縱性】

The Parity Ray Regularizer for Pacing in Auction Markets【用于拍賣市場節(jié)奏的 Parity Ray 正則化器】

推薦

冷啟動(dòng)

Alleviating Cold-start Problem in CTR Prediction with A Variational Embedding Learning Framework【使用變分嵌入學(xué)習(xí)框架緩解 CTR 預(yù)測中的冷啟動(dòng)問題】

PNMTA: A Pretrained Network Modulation and Task Adaptation Approach for User Cold-Start Recommendation【PMNTA：一種用于用戶冷啟動(dòng)推薦的預(yù)訓(xùn)練網(wǎng)絡(luò)調(diào)制和任務(wù)適應(yīng)方法】

KoMen: Domain Knowledge Guided Interaction Recommendation for Emerging Scenarios【KoMen：新興場景的領(lǐng)域知識引導(dǎo)交互推薦】

點(diǎn)擊CTR

Alleviating Cold-start Problem in CTR Prediction with A Variational Embedding Learning Framework【使用變分嵌入學(xué)習(xí)框架緩解 CTR 預(yù)測中的冷啟動(dòng)問題】

Deep Interest Highlight Network for Click-Through Rate Prediction in Trigger-Induced Recommendation【觸發(fā)誘導(dǎo)推薦中點(diǎn)擊率預(yù)測的深度興趣突出網(wǎng)絡(luò)】

ParClick: A Scalable Algorithm for EM-based Click Models【ParClick：基于 EM 的 Click 模型的可擴(kuò)展算法】

CBR: Context Bias aware Recommendation for Debiasing User Modeling and Click Prediction【CBR：用于消除用戶建模和點(diǎn)擊預(yù)測的上下文偏差感知建議】

Rating Distribution Calibration for Selection Bias Mitigation in Recommendations【推薦中減少選擇偏差的評級分布校準(zhǔn)】

MBCT: Tree-Based Feature-Aware Binning for Individual Uncertainty Calibration【MBCT：用于個(gè)體不確定性校準(zhǔn)的基于樹的特征感知分箱】

session會話推薦

Generative Session-based Recommendation【基于生成會話的推薦】

GSL4Rec: Session-based Recommendations with Collective Graph Structure Learning and Next Interaction Prediction【GSL4Rec：具有集體圖結(jié)構(gòu)學(xué)習(xí)和下一次交互預(yù)測的基于會話的推薦】

sequential序列推薦

Efficient Online Learning to Rank for Sequential Music Recommendation【高效的在線學(xué)習(xí)對順序音樂推薦進(jìn)行排名】

Filter-enhanced MLP is All You Need for Sequential Recommendation【過濾器增強(qiáng)的 MLP 是進(jìn)行順序推薦所需的全部】

Intent Contrastive Learning for Sequential Recommendation【序列推薦的意圖對比學(xué)習(xí)】

Learn from Past, Evolve for Future: Search-based Time-aware Recommendation with Sequential Behavior Data【從過去學(xué)習(xí)，為未來發(fā)展：基于搜索的時(shí)間感知推薦與順序行為數(shù)據(jù)】

Sequential Recommendation via Stochastic Self-Attention【通過隨機(jī)自注意力的順序推薦】

Sequential Recommendation with Decomposed Item Feature Routing【具有分解項(xiàng)目特征路由的順序推薦】

Towards Automatic Discovering of Deep Hybrid Network Architecture for Sequential Recommendation【面向順序推薦的深度混合網(wǎng)絡(luò)架構(gòu)的自動(dòng)發(fā)現(xiàn)】

Unbiased Sequential Recommendation with Latent Confounders【具有潛在混雜因素的無偏順序推薦】

Disentangling Long and Short-Term Interests for Recommendation【解耦推薦的長期和短期利益】

跨域

Collaborative Filtering with Attribution Alignment for Review-based Non-overlapped Cross Domain Recommendation【用于基于評論的非重疊跨域推薦的具有屬性對齊的協(xié)同過濾】

Differential Private Knowledge Transfer for Privacy-Preserving Cross-Domain Recommendation【隱私保護(hù)跨域推薦的差分私有知識遷移】

Learning Robust Recommenders through Cross-Model Agreement【通過跨模型協(xié)議學(xué)習(xí)強(qiáng)大的推薦器】

糾偏

Cross Pairwise Ranking for Unbiased Item Recommendation【無偏項(xiàng)目推薦的交叉成對排名】

CBR: Context Bias aware Recommendation for Debiasing User Modeling and Click Prediction【CBR：用于消除用戶建模和點(diǎn)擊預(yù)測的上下文偏差感知建議】

Unbiased Sequential Recommendation with Latent Confounders【具有潛在混雜因素的無偏順序推薦】

Asymptotically Unbiased Estimation for Delayed Feedback Modeling via Label Correction【通過標(biāo)簽校正的延遲反饋建模的漸近無偏估計(jì)】

UKD: Debiasing Conversion Rate Estimation via Uncertainty-regularized Knowledge Distillation【UKD：通過不確定性正則化知識蒸餾的去偏轉(zhuǎn)換率估計(jì)】

個(gè)性化

Regulatory Instruments for Fair Personalized Pricing【公平個(gè)性化定價(jià)的監(jiān)管工具】

Discovering Personalized Semantics for Soft Attributes in Recommender Systems Using Concept Activation Vectors【使用概念激活向量發(fā)現(xiàn)推薦系統(tǒng)中軟屬性的個(gè)性化語義】

Improving Personalized Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks【通過調(diào)整輔助任務(wù)的梯度幅度來改進(jìn)個(gè)性化推薦】

An Empirical Investigation of Personalization Factors on TikTok【TikTok個(gè)性化因素的實(shí)證研究】

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于圖神經(jīng)網(wǎng)絡(luò)的個(gè)性化可視化推薦】

因果關(guān)系

A Model-Agnostic Causal Learning Framework for Recommendation using Search Data【使用搜索數(shù)據(jù)進(jìn)行推薦的與模型無關(guān)的因果學(xué)習(xí)框架】

Causal Preference Learning for Out-of-Distribution Recommendation【分布外推薦的因果偏好學(xué)習(xí)】

Unbiased Sequential Recommendation with Latent Confounders【具有潛在混雜因素的無偏順序推薦】

可解釋

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于圖神經(jīng)網(wǎng)絡(luò)的個(gè)性化可視化推薦】

Path Language Modeling over Knowledge Graphs for Explainable Recommendation【可解釋推薦的知識圖路徑語言建模】

Accurate and Explainable Recommendation via Review Rationalization【通過審查合理化提供準(zhǔn)確且可解釋的建議】

AmpSum: Adaptive Multiple-Product Summarization towards Improving Recommendation Explainability【AmpSum：提高推薦可解釋性的自適應(yīng)多產(chǎn)品總結(jié)】

Comparative Explanations of Recommendations【推薦系統(tǒng)的比較解釋】

Neuro-Symbolic Interpretable Collaborative Filtering for Attribute-based Recommendation【基于屬性推薦的神經(jīng)符號可解釋協(xié)同過濾】

公平性

FairGAN: GANs-based Fairness-aware Learning for Recommendations with Implicit Feedback【FairGAN：基于 GAN 的公平感知學(xué)習(xí)，用于具有隱式反饋的推薦】

Regulatory Instruments for Fair Personalized Pricing【公平個(gè)性化定價(jià)的監(jiān)管工具】

Differential Private Knowledge Transfer for Privacy-Preserving Cross-Domain Recommendation【隱私保護(hù)跨域推薦的差分私有知識遷移

多任務(wù)

A Multi-task Learning Framework for Product Ranking with BERT【使用 BERT 進(jìn)行產(chǎn)品排名的多任務(wù)學(xué)習(xí)框架】

A Contrastive Sharing Model for Multi-Task Recommendation【多任務(wù)推薦的對比共享模型】

對比學(xué)習(xí)

Intent Contrastive Learning for Sequential Recommendation【序列推薦的意圖對比學(xué)習(xí)】

A Contrastive Sharing Model for Multi-Task Recommendation【多任務(wù)推薦的對比共享模型】

Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning【使用鄰域豐富的對比學(xué)習(xí)改進(jìn)圖協(xié)同過濾】

圖學(xué)習(xí)

Hypercomplex Graph Collaborative Filtering【超復(fù)雜圖協(xié)同過濾】

Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning【使用鄰域豐富的對比學(xué)習(xí)改進(jìn)圖協(xié)同過濾】

STAM: A Spatiotemporal Aggregation Method for Graph Neural Network-based Recommendation【STAM：一種基于圖神經(jīng)網(wǎng)絡(luò)推薦的時(shí)空聚合方法】

FIRE: Fast Incremental Recommendation with Graph Signal Processing【FIRE：使用圖信號處理的快速增量推薦】

Graph Neural Transport Networks with Non-local Attentions for Recommender Systems【用于推薦系統(tǒng)的具有非局部注意力的圖神經(jīng)傳輸網(wǎng)絡(luò)】

Multi-level Recommendation Reasoning over Knowledge Graphs with Reinforcement Learning【基于強(qiáng)化學(xué)習(xí)的知識圖的多級推薦推理】

Revisiting Graph Neural Network based Social Recommendation【重新審視基于圖神經(jīng)網(wǎng)絡(luò)的社交推薦】

VisGNN: Personalized Visualization Recommendation via Graph Neural Networks【VisGNN：基于圖神經(jīng)網(wǎng)絡(luò)的個(gè)性化可視化推薦】

Path Language Modeling over Knowledge Graphs for Explainable Recommendation【可解釋推薦的知識圖路徑語言建模】

GSL4Rec: Session-based Recommendations with Collective Graph Structure Learning and Next Interaction Prediction【GSL4Rec：具有集體圖結(jié)構(gòu)學(xué)習(xí)和下一次交互預(yù)測的基于會話的推薦】

Optimizing Rankings for Recommendation in Matching Markets【優(yōu)化匹配市場中推薦的排名】Yi Su, Magd Bayoumi and Thorsten Joachims

采樣

Learning Recommenders for Implicit Feedback with Importance Resampling【通過重要性重采樣學(xué)習(xí)隱式反饋的推薦器】

A Gain-Tuning Dynamic Negative Sampler for Recommendation【用于推薦的增益調(diào)整動(dòng)態(tài)負(fù)采樣器】Qiannan Zhu, Haobo Zhang, Qing He and Zhicheng Dou

新聞

FeedRec: News Feed Recommendation with Various User Feedbacks【FeedRec：具有各種用戶反饋的新聞提要推薦】

MINDSim: User Simulator for News Recommenders【MINDSim：新聞推薦者的用戶模擬器】

在線學(xué)習(xí)

Learning Neural Ranking Models Online from Implicit User Feedback【從隱式用戶反饋在線學(xué)習(xí)神經(jīng)排序模型】

Efficient Online Learning to Rank for Sequential Music Recommendation【高效的在線學(xué)習(xí)對順序音樂推薦進(jìn)行排名】

強(qiáng)化學(xué)習(xí)

Multiple Choice Questions based Multi-Interest Policy Learning for Conversational Recommendation【基于多項(xiàng)選擇題的會話推薦多興趣策略學(xué)習(xí)】

Off-policy Learning over Heterogeneous Information for Recommendation【用于推薦的異構(gòu)信息的異策略學(xué)習(xí)】

Multi-level Recommendation Reasoning over Knowledge Graphs with Reinforcement Learning【基于強(qiáng)化學(xué)習(xí)的知識圖的多級推薦推理】

其他

Learning Probabilistic Box Embeddings for Effective and Efficient Ranking【學(xué)習(xí)用于有效和高效排名的概率框嵌入】

Conditional Generation Net for Medication Recommendation【藥物推薦的條件生成網(wǎng)絡(luò)】

Automating Feature Selection in Deep Recommender Systems【深度推薦系統(tǒng)中的自動(dòng)特征選擇】

Choice of Implicit Signal Matters: Accounting for UserAspirations in Podcast Recommendations【隱式信號的選擇很重要：考慮播客推薦中的用戶愿望】

Deep Unified Representation for Heterogeneous Recommendation【異構(gòu)推薦的深度統(tǒng)一表示】

Learning to Augment for Casual User Recommendation【學(xué)習(xí)增強(qiáng)臨時(shí)用戶推薦】

Modality Matches Modality: Pretraining Modality-Disentangled Item Representations for Recommendation【模態(tài)匹配模態(tài)：預(yù)訓(xùn)練模態(tài)分離商品表征以進(jìn)行推薦】

Mutually-Regularized Dual Collaborative Variational Auto-encoder for Recommendation Systems【推薦系統(tǒng)的相互正則化雙協(xié)同變分自動(dòng)編碼器】

Re4: Learning to Re-contrast, Re-attend, Re-construct for Multi-interest Recommendation【Re4：學(xué)習(xí)重新對比、重新參與、重新構(gòu)建多興趣推薦】

Consensus Learning from Heterogeneous Objectives for One-Class Collaborative Filtering【一類協(xié)同過濾的異構(gòu)目標(biāo)的共識學(xué)習(xí)】

Fast Variational AutoEncoder with Inverted Multi-Index for Collaborative Filtering【用于協(xié)同過濾的具有倒置多索引的快速變分自動(dòng)編碼器】

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization【HRCF：通過雙曲幾何正則化增強(qiáng)協(xié)同過濾】

MCL: Mixed-Centric Loss for Collaborative Filtering【MCL：用于協(xié)同過濾的混合中心損失】

Stochastic-Expert Variational Autoencoder for Collaborative Filtering【用于協(xié)同過濾的隨機(jī)專家變分自動(dòng)編碼器】

Rewiring what-to-watch-next Recommendations to Reduce Radicalization Pathways【重新制定下一步觀看的建議以減少激進(jìn)化途徑】

Recommendation Unlearning

What to Watch Next: Two-side Interactive Networksfor Live Broadcast Recommendation【接下來看什么：直播推薦的雙邊互動(dòng)網(wǎng)絡(luò)】

參考

總結(jié)

以上是生活随笔為你收集整理的WWW 2022 推荐系统和广告相关论文整理分类的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。