作者｜牟善磊

?學校｜中國人民大學碩士生

研究方向 | 推薦系統

來源 |?RUC AI Box

推薦系統作為深度學習御三家（CV, NLP, RS）之一，一直都是學術界和工業界的熱門研究 topic。為了更加清楚的掌握推薦系統的前沿方向與最新進展，本文整理了最近一年頂會中推薦系統相關的論文，一共涵蓋 SIGIR2020, KDD2020, RecSys2020, CIKM2020, AAAI2021, WSDM2021, WWW2021 七個會議共 221 篇論文。本次整理以 long paper 和 research paper 為主，也包含少量的 short paper 和 industry paper。

本文整理的論文列表已經同步更新到 GitHub，GitHub 上會持續更新頂會論文，歡迎大家關注和 star。

https://github.com/RUCAIBox/Awesome-RSPapers

同時歡迎大家使用 RecBole（伯樂）推薦系統工具包。RecBole 是一個基于 PyTorch 實現的，面向研究者的，易于開發與復現的，統一、全面、高效的推薦系統代碼庫。目前已經有 72 個常見的推薦模型，覆蓋主流的研究任務，支持全面多樣的數據處理與評測方式。同時我們也整理了 28 個常用的推薦系統數據集，可以直接在 RecBole 中使用。

https://github.com/RUCAIBox/RecBole

https://github.com/RUCAIBox/RecSysDatasets

本文按照個人閱讀習慣和文章的側重點將這些論文分為以下五大類和若干小類：

?· 1 推薦任務·?

• Collaborative Filtering
  

• Sequential/Session-based Recommendations

• Knowledge-aware Recommendations

• Feature Interactions

• Conversational Recommender System

• Social Recommendations

• News Recommendations

• Text-aware Recommendations

• Point-of-Interest

• Online Recommendations

• Group Recommendations

• Multi-task/Multi-behavior/Cross-domain Recommendations

• Other Task

?· 2 推薦的熱門研究話題·?

• Debias in Recommender System
  

• Fairness in Recommender System

• Attack in Recommender System

• Explanation in Recommender System

• Long-tail/Cold-start in Recommender System

• Evaluation

?· 3 先進技術在推薦中的應用·?

• Pre-training in Recommender System
  

• Reinforcement Learning in Recommender System

• Knowledge Distillation in Recommender System

• NAS in Recommender System

• Federated Learning in Recommender System

?· 4 理論/實驗分析·?

?· 5 其他·?

01

推薦任務

1.1 Collaborative Filtering

協同過濾永不過時！作為經典中的經典，協同過濾更多考慮用戶和物品的交互信息來找到相似的用戶和物品，再利用相似性來進行推薦。從最早的基于鄰居（Neighbor）的方法 ItemKNN, UserKNN 等，到基于矩陣分解（MF）的方法 SVD++, BPRMF 等，再隨著神經網絡的發展，基于神經網絡（NN）的方法 NCF, CDAE 等，再到最近一些基于圖神經網絡的方法，協同過濾一直在不斷進步，也展現出越來越強大的威力。

最新的一些研究工作主要集中在如何更好的捕捉相似性上，包括設計更為復雜的網絡結構，主要以 VAE，GNN 的結構為主；更復雜的空間來度量，例如雙曲空間，立方體表示；以及更復雜的相似度計算方式。

1. LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation. SIGIR 2020 【GNN-based】

2. Deep Critiquing for VAE-based Recommender Systems. SIGIR 2020 【VAE-based】

3. Neighbor Interaction Aware Graph Convolution Networks for Recommendation. SIGIR 2020 【GNN-based】

4. A Framework for Recommending Accurate and Diverse Items Using Bayesian Graph Convolutional Neural Networks. KDD 2020 【GNN-based】

5. Dual Channel Hypergraph Collaborative Filtering. KDD 2020 【基于超圖的CF】

6. Probabilistic Metric Learning with Adaptive Margin for Top-K Recommendation. KDD 2020 【新的度量方式】

7. Content-Collaborative Disentanglement Representation Learning for Enhanced Recommendation. RecSys 2020

8. Neural Collaborative Filtering vs. Matrix Factorization Revisited. RecSys 2020 【Rendle的文章，分析MF在效果和效率上都要比NCF好】

9. Bilateral Variational Autoencoder for Collaborative Filtering. WSDM 2021【user，item雙向VAE】

10. Learning User Representations with Hypercuboids for Recommender Systems. WSDM 2021 【立方體表示】

11. Local Collaborative Autoencoders. WSDM 2021

12. A Scalable, Adaptive and Sound Nonconvex Regularizer for Low-rank Matrix Completion. WWW 2021 【一種低秩矩陣分解方法】

13. HGCF: Hyperbolic Graph Convolution Networks for Collaborative Filtering. WWW 2021 【雙曲圖卷積用于CF，雙曲卷積已經有不少工作了，也出現在RS了】

14. High-dimensional Sparse Embeddings for Collaborative Filtering. WWW 2021 【高維稀疏表示】

15. Collaborative Filtering with Preferences Inferred from Brain Signals. WWW 2021 【Brain Signals】

16. Interest-aware Message-Passing GCN for Recommendation. WWW 2021

17. Neural Collaborative Reasoning. WWW 2021 【用NN建模logical reasoning來代替inner product】

18. Sinkhorn Collaborative Filtering. WWW 2021

19. Disentangling User Interest and Conformity for Recommendation with Causal Embedding. WWW 2021

1.2 Sequential/Session-based Recommendations

序列推薦也是頂會中熱門的研究方向。sequential recommendations 和 session-based recommendations 嚴格來說并不完全相似，與 sequential recommendations 相比，session-based recommendations 不太有 user 的概念以及序列更短，但兩者有非常多的相似之處，都是通過交互過的 item 序列來刻畫偏好，因此放在一起來說。

序列推薦最早是基于 Markov Chain 來進行建模，例如經典方法 FPMC，后來隨著一些序列模型的發展，例如 RNN，GRU，包括最近火熱的 Transformer，序列推薦開始基于這些序列模型來建模，例如 GRU4Rec，SASRec 等，隨著圖神經網絡的發展，基于 GNN 的方法也變得非常熱門。

1. Sequential Recommendation with Self-attentive Multi-adversarial Network. SIGIR 2020

2. KERL: A Knowledge-Guided Reinforcement Learning Model for Sequential Recommendation. SIGIR 2020 【SR + KG + RL】

3. Modeling Personalized Item Frequency Information for Next-basket Recommendation. SIGIR 2020 【融合item頻率】

4. Incorporating User Micro-behaviors and Item Knowledge into Multi-task Learning for Session-based Recommendation. SIGIR 2020【SR + KG + MTL】

5. GAG: Global Attributed Graph Neural Network for Streaming Session-basedRecommendation. SIGIR 2020 【Streaming SR】

6. Next-item Recommendation with Sequential Hypergraphs. SIGIR 2020 【基于超圖】

7. A General Network Compression Framework for Sequential Recommender Systems. SIGIR 2020 【通用的SR模型壓縮方法】

8. Make It a Chorus: Knowledge- and Time-aware Item Modeling for Sequential Recommendation. SIGIR 2020 【SR + KG】

9. Global Context Enhanced Graph Neural Networks for Session-based Recommendation. SIGIR 2020

10. Self-Supervised Reinforcement Learning for Recommender Systems. SIGIR 2020 【SR + RL】

11. Time Matters: Sequential Recommendation with Complex Temporal Information. SIGIR 2020 【Time-aware】

12. Controllable Multi-Interest Framework for Recommendation. SIGIR 2020 【多興趣】

13. Disentangled Self-Supervision in Sequential Recommenders. KDD 2020 【多興趣】

14. Handling Information Loss of Graph Neural Networks for Session-based Recommendation. KDD 2020

15. Contextual and Sequential User Embeddings for Large-Scale Music Recommendation. RecSys 2020 【針對music場景】

16. FISSA:Fusing Item Similarity Models with Self-Attention Networks for Sequential Recommendation. RecSys 2020 【SASRec的基礎上融合物品相似性】

17. From the lab to production: A case study of session-based recommendations in the home-improvement domain. RecSys 2020 【對SR方法評測分析】

18. Recommending the Video to Watch Next: An Offline and Online Evaluation at YOUTV.de. RecSys 2020 【基于Markov Chain】

19. SSE-PT:Sequential Recommendation Via Personalized Transformer. RecSys 2020 【改進SASRec】

20. Improving End-to-End Sequential Recommendations with Intent-aware Diversification. CIKM 2020

21. Quaternion-based self-Attentive Long Short-term User Preference Encoding for Recommendation. CIKM 2020 【Quaternion捕捉長短期興趣】

22. Sequential Recommender via Time-aware Attentive Memory Network. CIKM 2020 【Time-aware】

23. Star Graph Neural Networks for Session-based Recommendation. CIKM 2020

24. Dynamic Memory Based Attention Network for Sequential Recommendation. AAAI 2021

25. Noninvasive Self-Attention for Side Information Fusion in Sequential Recommendation. AAAI 2021【BERT + Context】

26. Self-Supervised Hypergraph Convolutional Networks for Session-based Recommendation. AAAI 2021 【基于超圖】

27. An Efficient and Effective Framework for Session-based Social Recommendation. WSDM 2021

28. Sparse-Interest Network for Sequential Recommendation. WSDM 2021 【多興趣】

29. Dynamic Embeddings for Interaction Prediction. WWW 2021

30. Session-aware Linear Item-Item Models for Session-based Recommendation. WWW 2021

31. RetaGNN: Relational Temporal Attentive Graph Neural Networks for Holistic Sequential Recommendation. WWW 2021

32. Adversarial and Contrastive Variational Autoencoder for Sequential Recommendation. WWW 2021 【VAE】

33. Future-Aware Diverse Trends Framework for Recommendation. WWW 2021 【融合特征】

34. Linear-Time Self Attention with Codeword Histogram for Efficient Recommendation. WWW 2021 【線性時間的selfattention】

35. DeepRec: On-device Deep Learning for Privacy-Preserving Sequential? Recommendation in MobileCommerce. WWW 2021 【序列推薦中的隱私保護】

1.3 Knowledge-aware Recommendations

這一部分主要包括利用結構化信息來幫助推薦系統的工作，用到的結構化信息主要以 Knowledgegraph 和 HIN 為主。這一類推薦任務也是近些年來熱門的研究方向。為了能充分發揮出結構化信息的優勢，早期有不少基于 meta-path 的方法，隨著圖神經網絡的發展，目前 GNN 的方法已經成為主流。

1. CKAN: Collaborative Knowledge-aware Attentive Network for Recommender Systems. SIGIR 2020

2. Attentional Graph Convolutional Networks for Knowledge Concept Recommendation in MOOCs in a Heterogeneous View. SIGIR2020 【MOOC推薦】

3. MVIN: Learning multiview items for recommendation. SIGIR 2020

4. Jointly Non-Sampling Learning for Knowledge Graph Enhanced Recommendation. SIGIR 2020

5. Joint Item Recommendation and Attribute Inference: An Adaptive Graph Convolutional Network Approach. SIGIR 2020

6. Leveraging Demonstrations for Reinforcement Recommendation Reasoning over Knowledge Graphs. SIGIR 2020 【RL】

7. SimClusters Community-Based Representations for Heterogenous Recommendations at Twitter. KDD 2020 【IndustryPaper by Twitter】

8. Multi-modal Knowledge Graphs for Recommender Systems. CIKM 2020

9. DisenHAN Disentangled Heterogeneous Graph Attention Network for Recommendation. CIKM 2020

10. Genetic Meta-Structure Search for Recommendation on Heterogeneous Information Network. CIKM 2020 【自動優化meta-path】

11. TGCN Tag Graph Convolutional Network for Tag-Aware Recommendation. CIKM 2020

12. Knowledge-Enhanced Top-K Recommendation in Poincaré Ball. AAAI 2021

13. Graph Heterogeneous Multi-Relational Recommendation. AAAI 2021

14. Knowledge-Enhanced Hierarchical Graph Transformer Network for Multi-Behavior Recommendation. AAAI 2021

15. Alleviating Cold-Start Problems in Recommendation through Pseudo-Labelling over Knowledge Graph. WSDM 2021

16. Decomposed Collaborative Filtering Modeling Explicit and Implicit Factors For Recommender Systems. WSDM 2021

17. Temporal Meta-path Guided Explainable Recommendation. WSDM 2021

18. Learning Intents behind Interactions with Knowledge Graph for Recommendation. WWW 2021

1.4 Feature Interactions

這一部分工作主要以特征交互為主。已經有非常多經典的方法在工業界的推薦中得到了廣泛的應用，例如 FM，DeepFM，Wide&Deep 等。

1. Detecting Beneficial Feature Interactions for Recommender Systems. AAAI 2021 【Graph-based】

2. DeepLight: Deep Lightweight Feature Interactions for Accelerating CTR Predictions in Ad Serving. WSDM 2021 【加速特征交互】

3. Multi-Interactive Attention Network for Fine-grained Feature Learning in CTR Prediction. WSDM 2021 【細粒度】

4. FM^2: Field-matrixed Factorization Machines for CTR Prediction. WWW 2021 【FwFM升級版】

1.5 Conversational Recommender System

1. Towards Question-based Recommender Systems.SIGIR 2020

2. Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion. KDD 2020 【CRS + KG】

3. Interactive Path Reasoning on Graph for Conversational Recommendation. KDD 2020

4. A Ranking Optimization Approach to Latent Linear Critiquing for Conversational Recommender Systems. RecSys 2020 【對話推薦系統的一種排序優化方法】

5. What does BERT know about books, movies and music：Probing BERT for Conversational Recommendation.RecSys 2020 【CRS + KG】

6. Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation. WSDM 2021 【多輪】

7. A Workflow Analysis of Context-driven Conversational Recommendation. WWW 2021 【分析CRS workflow】

1.6 Social Recommendations

1. Partial Relationship Aware Influence Diffusion via a Multi-channel Encoding Scheme for Social Recommendation. CIKM2020

2. Random Walks with Erasure: Diversifying Personalized Recommendations on Social and Information Networks. WWW 2021

3. Dual Side Deep Context-aware Modulation for Social Recommendation. WWW 2021

4. Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation. WWW 2021

1.7 News Recommendations

1. KRED: Knowledge-Aware Document Representation for News Recommendations. RecSys 2020

2. News Recommendation with Topic-Enriched Knowledge Graphs. CIKM 2020

3. The Interaction between Political Typology and Filter Bubbles in News Recommendation Algorithms. WWW 2021

1.8 Text-aware Recommendations

1. TAFA: Two-headed Attention Fused Autoencoder for Context-Aware Recommendations. RecSys 2020 【Review】

2. Set-Sequence-Graph A Multi-View Approach Towards Exploiting Reviews for Recommendation. CIKM 2020 【Review】

3. TPR: Text-aware Preference Ranking for Recommender Systems. CIKM 2020

4. Leveraging Review Properties for Effective Recommendation. WWW 2021 【Review】

1.9 Point-of-Interest

1. HME: A Hyperbolic Metric Embedding Approach for Next-POI Recommendation. SIGIR 2020

2. Spatial Object Recommendation with Hints: When Spatial Granularity Matters. SIGIR 2020

3. Geography-Aware Sequential Location Recommendation. KDD 2020

4. Learning Graph-Based Geographical Latent Representation for Point-of-Interest Recommendation. CIKM 2020

5. STP-UDGAT Spatial-Temporal-Preference User Dimensional Graph Attention Network for Next POI Recommendation. CIKM 2020

6. STAN: Spatio-Temporal Attention Network for next Point-of-Interest Recommendation. WWW 2021

7. Incremental Spatio-Temporal Graph Learning for Online Query-POI Matching. WWW 2021

1.10 Online Recommendations

1. Gemini: A novel and universal heterogeneous graph information fusing framework for online recommendations. KDD 2020

2. Maximizing Cumulative User Engagement in Sequential Recommendation An Online Optimization Perspective. KDD 2020

3. Exploring Clustering of Bandits for Online Recommendation System. RecSys 2020

4. Contextual User Browsing Bandits for Large-Scale Online Mobile Recommendation. RecSys 2020

5. A Hybrid Bandit Framework for Diversified Recommendation. AAAI 2021

1.11Group Recommendations

1. GAME: Learning Graphical and Attentive Multi-view Embeddings for Occasional Group Recommendation. SIGIR 2020

2. GroupIM: A Mutual Information Maximizing Framework for Neural Group Recommendation. SIGIR 2020

3. Group-Aware Long- and Short-Term Graph Representation Learning for Sequential Group Recommendation. SIGIR 2020

1.12 Multi-task/Multi-behavior/Cross-domain

1. Transfer Learning via Contextual Invariants for One-to-Many Cross-Domain Recommendation. SIGIR 2020 【Cross-domain】

2. CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network. SIGIR 2020 【Cross-domain】

3. Multi-behavior Recommendation with Graph Convolution Networks. SIGIR 2020

4. Parameter-Efficient Transfer from Sequential Behaviors for User Modeling and Recommendation. SIGIR 2020

5. Web-to-Voice Transfer for Product Recommendation on Voice. SIGIR 2020

6. Jointly Learning to Recommend andAdvertise. KDD 2020

7. Progressive Layered Extraction (PLE) A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations. RecSys2020 【多目標優化 by Tencent】

8. Whole-Chain Recommendations. CIKM 2020

9. Personalized Approximate Pareto-Efficient Recommendation. WWW 2021 【面向Pareto的強化學習方法解決多目標優化推薦】

1.13 Other Task

除了上面提到的比較經典的推薦任務，最近一年的頂會文章還有一些其他有意思的任務。

1. Hierarchical Fashion Graph Network for Personalized Outfit Recommendation. SIGIR 2020 【OutfitRecommendation】

2. Octopus: Comprehensive and Elastic User Representation for the Generation of Recommendation Candidates. SIGIR 2020 【Candidate Generation】

3. Goal-driven Command Recommendations for Analysts. RecSys 2020 【從非結構化log數據中進行command recommendation】

4. MultiRec: A Multi-Relational Approach for Unique Item Recommendation in Auction Systems.?RecSys 2020 【拍賣系統中的推薦】

5. PURS: Personalized Unexpected Recommender System for Improving User Satisfaction. RecSys 2020 【意外推薦，沒想到吧.jpg】

6. RecSeats: A Hybrid Convolutional Neural Network Choice Model for Seat 7. Recommendations at Reserved Seating Venues.RecSys 2020 【座位推薦，萊納你坐啊】

8. Live Multi-Streaming and Donation Recommendations via Coupled Donation-Response Tensor Factorization. CIKM 2020 【Multi-streaming】

9. Learning to Recommend from Sparse Data via Generative User Feedback. AAAI 2021

10. Real-time Relevant Recommendation Suggestion. WSDM 2021 【Recommendation Suggestion】

11. Heterogeneous Graph Augmented Multi-Scenario Sharing Recommendation with Tree-Guided Expert Networks. WSDM2021 【Share Recommendation】

12. FINN: Feedback Interactive Neural Network for Intent Recommendation. WWW 2021 【IntentRecommendation】

13. Drug Package Recommendation via Interaction-aware Graph Induction. WWW 2021 【DrugPackage Recommendation】

14. Large-scale Comb-K Recommendation. WWW 2021【Comb-K 推薦】

15. Variation Control and Evaluation for Generative Slate Recommendations. WWW 2021 【GenerativeSlate Recommendation】

16. Diversified Recommendation Through Similarity-Guided Graph Neural Networks. WWW 2021 【多樣性推薦】

02

推薦的熱門研究話題

2.1 Debias in Recommender System

bias 是廣泛存在于推薦系統中的，包括比較常見的流行度偏差 (popularity bias)，選擇偏差 (selection bias)，曝光偏差 (exposure bias)，位置偏差 (position bias) 以及其他各種偏差。推薦系統的 debias 的工作一直都有人在做，隨著近兩年 causal inference 成為熱點，這個 topic 最近變得非常熱門。

1. A General Knowledge Distillation Framework for Counterfactual Recommendation via Uniform Data. SIGIR 2020

2. Measuring and Mitigating Item Under-Recommendation Bias in Personalized Ranking Systems. SIGIR 2020

3. Attribute-based Propensity for Unbiased Learning in Recommender Systems Algorithm and Case Studies. KDD 2020 【position bias】

4. Counterfactual Evaluation of Slate Recommendations with Sequential Reward Interactions. KDD 2020

5. Debiasing Item-to-Item Recommendations With Small Annotated Datasets. RecSys 2020

6. Keeping Dataset Biases out of the Simulation : A Debiased Simulator for Reinforcement Learning based RecommenderSystems. RecSys 2020

7. Unbiased Ad Click Prediction for Position-aware Advertising Systems. RecSys 2020 【debiasin position-aware recommedantion】

8. Unbiased Learning for the Causal Effect of Recommendation. RecSys 2020

9. E-commerce Recommendation with Weighted Expected Utility. CIKM 2020

10. Popularity-Opportunity Bias in Collaborative Filtering. WSDM 2021 【提出了一種新的popularitybias考慮了opportunity】

11. Combating Selection Biases in Recommender Systems with a Few Unbiased Ratings. WSDM 2021 【利用少部分無偏數據解決selection bias】

12. Leave No User Behind Towards Improving the Utility of Recommender Systems for Non-mainstream Users. WSDM 2021 【mainstream bias】

13. Non-Clicks Mean Irrelevant Propensity Ratio Scoring As a Correction. WSDM 2021

14. Diverse User Preference Elicitation with Multi-Armed Bandits. WSDM 2021 【多臂老虎機增加推薦系統多樣性緩解popularitybias】

15. Unbiased Learning to Rank in Feeds Recommendation. WSDM 2021 【context-aware position bias】

16. Cross-Positional Attention for Debiasing Clicks. WWW 2021

17. Debiasing Career Recommendations with Neural Fair Collaborative Filtering. WWW 2021 【genderbias】

2.2 Fairness in Recommender System

1. airness-Aware Explainable Recommendation over Knowledge Graphs. SIGIR 2020 【運用KG做re-ranking】

2. Ensuring Fairness in Group Recommendations by Rank-Sensitive Balancing of Relevance. RecSys 2020

3. Fairness-Aware News Recommendation with Decomposed Adversarial Learning. AAAI 2021 【Fairness inNews Recommendation】

4. Practical Compositional Fairness Understanding Fairness in Multi-Component Recommender Systems. WSDM 2021

5. Towards Long-term Fairness in Recommendation. WSDM 2021

6. Learning Fair Representations for Recommendation: A Graph-based Perspective. WWW 2021

7. User-oriented Group Fairness In Recommender Systems. WWW 2021

2.3 Attack in Recommender System

萬物皆可攻擊：

1. Revisiting Adversarially Learned Injection Attacks Against Recommender Systems. RecSys 2020 【對推薦系統對抗性注入攻擊的重新審視】

2. Attacking Recommender Systems with Augmented User Profiles. CIKM 2020 【從用戶特征進行攻擊】

3. A Black-Box Attack Model for Visually-Aware Recommenders. WSDM 2021 【針對圖像特征的推薦系統攻擊】

4. Denoising Implicit Feedback for Recommendation. WSDM 2021 【訓練數據去噪】

5. Adversarial Item Promotion: Vulnerabilities at the Core of Top-N 6. Recommenders that Use Images to Address Cold Start. WWW2021 【針對圖像攻擊】

6. Graph Embedding for Recommendation against Attribute Inference Attacks. WWW 2021

2.4 Explanation in Recommender System

1. Try This Instead: Personalized and Interpretable Substitute Recommendation. KDD 2020

2. CAFE: Coarse-to-Fine Neural Symbolic Reasoning for Explainable Recommendation. CIKM 2020 【通過KG提升可解釋性】

3. Explainable Recommender Systems via Resolving Learning Representations. CIKM 2020 【借助KG表示學習提升可解釋性】

4. Generate Neural Template Explanations for Recommendation. CIKM 2020 【生成文本提升可解釋性】

5. Explainable Recommendation with Comparative Constraints on Product Aspects. WSDM 2021 【通過和物品在屬性上比較來提升可解釋性】

6. Explanation as a Defense of Recommendation. WSDM 2021 【生成文本提升可解釋性】

7. EX^3: Explainable Product Set Recommendation for Comparison Shopping. WWW 2021

8. Learning from User Feedback on Explanations to Improve Recommender Models. WWW 2021

2.5 Long-tail/Cold-start in Recommender System

從有推薦系統開始就存在的痛點問題，目前比較流行基于 meta-learning, transfer-learning 的方法。

1. Content-aware Neural Hashing for Cold-start Recommendation. SIGIR 2020

2. MAMO: Memory-Augmented Meta-Optimization for Cold-start Recommendation. KDD 2020

3. Learning Transferrable Parameters for Long-tailed Sequential User Behavior Modeling. KDD 2020 【通用的方法解決序列推薦中的長尾問題】

4. Meta-learning on Heterogeneous Information Networks for Cold-start Recommendation. KDD 2020

5. Cold-Start Sequential Recommendation via Meta Learner. AAAI 2021

6. Personalized Adaptive Meta Learning for Cold-start User Preference Prediction. AAAI 2021

7. Task-adaptive Neural Process for User Cold-Start Recommendation. WWW 2021

8. A Model of Two Tales: Dual Transfer Learning Framework for Improved Long-tail Item Recommendation. WWW 2021

2.6 Evaluation

對于新型推薦任務評測方式的創造以及到底該不該采樣評測？

1. Measuring Recommendation Explanation Quality: The Conflicting Goals of Explanations. SIGIR 2020 【評價生成的推薦系統解釋的質量】

2. Evaluating Conversational Recommender Systems via User Simulation. KDD 2020 【CRS評測指標】

3. On Sampled Metrics for Item Recommendation.KDD 2020

4. On Sampling Top-K Recommendation Evaluation. KDD 2020

5. Are We Evaluating Rigorously:Benchmarking Recommendation for Reproducible Evaluation and FairComparison. RecSys 2020

6. On Target Item Sampling in Offline Recommender System Evaluation. RecSys 2020

03

先進技術在推薦中的應用

3.1 Pre-training in Recommender System

1. S^3-Rec: Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximization.? CIKM 2020

2. U-BERT Pre-Training User Representations for Improved Recommendation. AAAI 2021

3. Pre-Training Graph Neural Networks for Cold-Start Users and Items Representation. WSDM 2021

3.2 Reinforcement Learning in Recommender System

1. MaHRL: Multi-goals Abstraction based Deep Hierarchical Reinforcement Learning for Recommendations. SIGIR 2020

2. Interactive Recommender System via Knowledge Graph-enhanced Reinforcement Learning. SIGIR 2020

3. Joint Policy-Value Learning for Recommendation. KDD 2020

4. BLOB: A Probabilistic Model for Recommendation that Combines Organic and Bandit Signals. KDD 2020

5. Learning to Collaborate in Multi-Module Recommendation via Multi-Agent Reinforcement Learning without Communication.RecSys 2020

6. Reinforcement Learning with a Disentangled Universal Value Function for Item Recommendation. AAAI 2021

7. User Response Models to Improve a REINFORCE Recommender System. WSDM 2021

8. Cost-Effective and Interpretable Job Skill Recommendation with Deep Reinforcement Learning. WWW 2021

9. A Multi-Agent Reinforcement Learning Framework for Intelligent Electric Vehicle Charging Recommendation. WWW 2021

10. Reinforcement Recommendation with User Multi-aspect Preference. WWW 2021

3.3 Knowledge Distillation in Recommender System

1. Privileged Features Distillation at Taobao Recommendations. KDD 2020

2. DE-RRD: A Knowledge Distillation Framework for Recommender System. CIKM 2020 【BPRMF也需要蒸餾我是沒想到的】

3. Bidirectional Distillation for Top-K Recommender System. WWW 2021 【雙向蒸餾】

3.4 NAS in Recommender System

1. Neural Input Search for Large Scale Recommendation Models. KDD 2020

2. Field-aware Embedding Space Searching in Recommender Systems. WWW 2021 【AutoML自動選擇特征維度】

3.5 Federated Learning in Recommender System

1. FedFast Going Beyond Average for Faster Training of Federated Recommender Systems. KDD 2020

04

理論/實驗分析

推薦系統理論分析的論文不是很常見，下面這幾篇都還挺有趣的。

1. How Dataset Characteristics Affect the Robustness of Collaborative?Recommendation Models. SIGIR 2020

2. Agreement and Disagreement between True and False-Positive Metrics in Recommender Systems Evaluation. SIGIR 2020

3. Critically Examining the Claimed Value of Convolutions over User-Item Embedding Maps for Recommender Systems. CIKM 2020 【理論結合實驗分析CNN建模推薦系統embedding不太work】

4. On Estimating Recommendation Evaluation Metrics under Sampling. AAAI 2021

5. Beyond Point Estimate Inferring Ensemble Prediction Variation from Neuron Activation Strength in Recommender Systems.WSDM 2021 【分析推薦系統ensemble】

6. Bias-Variance Decomposition for Ranking. WSDM 2021

7. Theoretical Understandings of Product Embedding for E-commerce Machine Learning. WSDM 2021

05

其他

1. Learning Personalized Risk Preferences for Recommendation. SIGIR 2020

2. Distributed Equivalent Substitution Training for Large-Scale Recommender Systems. SIGIR 2020 【大規模推薦系統的分布式訓練】

3. Beyond User Embedding Matrix: Learning to Hash for Modeling Large-Scale Users in Recommendation. SIGIR 2020 【大規模user hash】

4. How to Retrain a Recommender System? SIGIR2020 【來了新數據怎么retrain】

5. Recommendation for New Users and New Items via Randomized Training and Mixture-of-Experts Transformation. SIGIR 2020

6. Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems. KDD 2020 【解決embedding內存問題】

7. Improving Recommendation Quality in Google Drive. KDD 2020 【Google Drive 推薦實戰】

8. A Method to Anonymize Business Metrics to Publishing Implicit Feedback Datasets. RecSys 2020 【如何構建和發布數據集】

9. Exploiting Performance Estimates for Augmenting Recommendation Ensembles. RecSys 2020 【有效的推薦模型ensemble方法】

10. User Simulation via Supervised Generative Adversarial Network. WWW 2021

????

現在，在「知乎」也能找到我們了

進入知乎首頁搜索「PaperWeekly」

點擊「關注」訂閱我們的專欄吧

關于PaperWeekly

PaperWeekly 是一個推薦、解讀、討論、報道人工智能前沿論文成果的學術平臺。如果你研究或從事 AI 領域，歡迎在公眾號后臺點擊「交流群」，小助手將把你帶入 PaperWeekly 的交流群里。

總結

以上是生活随笔為你收集整理的从200多篇顶会论文看推荐系统前沿方向与最新进展的全部內容，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站內容還不錯，歡迎將生活随笔推薦給好友。