|
| 1 | +# 推荐系统工业界顶会论文总结——KDD 2021 |
| 2 | + |
| 3 | +[知乎专栏](https://zhuanlan.zhihu.com/p/467276380) |
| 4 | + |
| 5 | + |
| 6 | +1. [Reinforced Anchor Knowledge Graph Generation for News Recommendation Reasoning](https://www.microsoft.com/en-us/research/uploads/prod/2021/05/KDD2021-anchorkg.pdf) |
| 7 | +Author(Institute): Jianxun Lian(Microsoft二作) |
| 8 | +KeyWords: news recommender; knowledge graph; recommendation reasoning |
| 9 | +Dataset: MIND; Bing News |
| 10 | + |
| 11 | +2. [Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender System](https://arxiv.org/pdf/2010.15363.pdf) |
| 12 | +Author(Institute): Jinfeng Yi(JD三作) |
| 13 | +KeyWords: Recommendation; Popularity Bias; Causal Reasoning |
| 14 | +Dataset: ML10M; Adressa; Globo; Gowalla; Yelp |
| 15 | + |
| 16 | +3. [Modeling the Sequential Dependence among Audience Multi-step Conversions with Multi-task Learning in Targeted Display Advertising](https://arxiv.org/pdf/2105.08489.pdf) |
| 17 | +Author(Institute): Dongbo Xi(Meituan) |
| 18 | +KeyWords: Sequential Dependence; Multi-step Conversions; Multi-task Learning; Targeted Display Advertising |
| 19 | +Dataset: Meituan; Co-Branded Credit Cards; Ali-CCP |
| 20 | + |
| 21 | +4. [Learning to Expand Audience via Meta Hybrid Experts and Critics for Recommendation and Advertising](https://arxiv.org/pdf/2105.14688.pdf) |
| 22 | +Author(Institute): Yudan Liu(WeChat三作) |
| 23 | +KeyWords: Look-alike; Audience Expansion; Meta Learning; Campaign |
| 24 | +Dataset: Tencent; WeChat |
| 25 | + |
| 26 | +5. [Adversarial Feature Translation for Multi-domain Recommendation](https://nlp.csai.tsinghua.edu.cn/~xrb/publications/KDD-2021_AFT.pdf) |
| 27 | +Author(Institute): Xiaobo Hao(WeChat) |
| 28 | +KeyWords: recommender system; multi-domain recommendation; GAN |
| 29 | +Dataset: Netflix; MDR-5B |
| 30 | + |
| 31 | +6. [Debiasing Learning based Cross-domain Recommendation](https://dl.acm.org/doi/abs/10.1145/3447548.3467067) |
| 32 | +Author(Institute): Yuxiao Dong(Alibaba二作) |
| 33 | +KeyWords: Debias |
| 34 | + |
| 35 | +7. [MixGCF: An Improved Training Method for Graph Neural Network-based Recommender Systems](https://keg.cs.tsinghua.edu.cn/jietang/publications/KDD21-Huang-et-al-MixGCF.pdf) |
| 36 | +Author(Institute): Yuxiao Dong(Facebook) |
| 37 | +KeyWords: Collaborative Filtering; Recommender Systems; Graph Neural Networks; Negative Samplin |
| 38 | +Dataset: Alibaba; Yelp2018; Amazon |
| 39 | + |
| 40 | +8. [Multi-view Denoising Graph Auto-Encoders on Heterogeneous Information Networks for Cold-start Recommendation](https://dl.acm.org/doi/10.1145/3447548.3467427) |
| 41 | +Author(Institute): Hao Gu(Tencent三作) |
| 42 | +KeyWords: Cold-start; Auto-Encoders; Denoise |
| 43 | +Dataset: WeChat |
| 44 | + |
| 45 | +9. [Reinforced Anchor Knowledge Graph Generation for News Recommendation Reasoning](https://www.microsoft.com/en-us/research/uploads/prod/2021/05/KDD2021-anchorkg.pdf) |
| 46 | +Author(Institute): Jianxun Lian(Microsoft二作) |
| 47 | +KeyWords: news recommender; knowledge graph; recommendation reasoning |
| 48 | +Dataset: MIND; Bing News |
| 49 | + |
| 50 | +10. [A Semi-Personalized System for User Cold Start Recommendation on Music Streaming Apps](https://arxiv.org/pdf/2106.03819.pdf) |
| 51 | +Author(Institute): Léa Briand(Deezer) |
| 52 | +KeyWords: Recommender Systems; User Cold Start; Music Streaming Service; Semi-Personalization; Heterogeneous Data; A/B Testing |
| 53 | +Dataset: Deezer |
| 54 | + |
| 55 | +11. [Architecture and Operation Adaptive Network for Online Recommendations](https://dl.acm.org/doi/10.1145/3447548.3467133) |
| 56 | +Author(Institute): Lang Lang(Didi Chuxing) |
| 57 | +KeyWords: Online Recommendations |
| 58 | + |
| 59 | +12. [SEMI: A Sequential Multi-Modal Information Transfer Network for E-Commerce Micro-Video Recommendations](https://dl.acm.org/doi/10.1145/3447548.3467189) |
| 60 | +Author(Institute): Chenyi Lei(Alibaba) |
| 61 | +KeyWords: E-Commerce Micro-Video Recommendation; Information Transfer |
| 62 | +Dataset: Taobao |
| 63 | + |
| 64 | +13. [Curriculum Meta-Learning for Next POI Recommendation](https://dl.acm.org/doi/abs/10.1145/3447548.3467132) |
| 65 | +Author(Institute): Miao Fan(Baidu三作) |
| 66 | +KeyWords: POI |
| 67 | +Dataset: Baidu Map |
| 68 | + |
| 69 | +14. [Learning to Embed Categorical Features without Embedding Tables for Recommendation](https://arxiv.org/pdf/2010.10784.pdf) |
| 70 | +Author(Institute): Wang-Cheng Kang(Google) |
| 71 | +KeyWords: Embed Categorical Features |
| 72 | +Dataset: Movielens20M; Amazon Book |
| 73 | + |
| 74 | +15. [Preference Amplification in Recommender Systems](https://dl.acm.org/doi/abs/10.1145/3447548.3467298) |
| 75 | +Author(Institute): Smriti Bhagat(Facebook二作) |
| 76 | +KeyWords: Recommender systems; echo chambers; filter bubbles; fixed point |
| 77 | +Dataset: MovieLens 10M; Yahoo |
| 78 | + |
| 79 | +16. [Data Poisoning Attack against Recommender System Using Incomplete and Perturbed Data](https://dl.acm.org/doi/abs/10.1145/3447548.3467233) |
| 80 | +Author(Institute): Yaliang Li(Alibaba三作) |
| 81 | +KeyWords: Attack |
| 82 | + |
| 83 | +17. [Initialization Matters: Regularizing Manifold-informed Initialization for Neural Recommendation Systems](https://arxiv.org/pdf/2106.04993.pdf) |
| 84 | +Author(Institute): Chunyan Miao(Alibaba三作) |
| 85 | +KeyWords: network initialization; recommender systems; manifold learning |
| 86 | +Dataset: ML-1M; Steam; Anime |
| 87 | + |
| 88 | +18. [We Know What You Want: An Advertising Strategy Recommender System for Online Advertising](https://arxiv.org/pdf/2105.14188.pdf) |
| 89 | +Author(Institute): Junqi Jin(Alibaba二作) |
| 90 | +KeyWords: E-commerce; Display Advertisement; Advertising Strategy Recommendation |
| 91 | +Dataset: online |
| 92 | + |
| 93 | +19. [A Unified Solution to Constrained Bidding in Online Display Advertising](https://dl.acm.org/doi/10.1145/3447548.3467199) |
| 94 | +Author(Institute): Yue He(Alibaba) |
| 95 | +KeyWords: advertising |
| 96 | +Dataset: Taobao |
| 97 | + |
| 98 | +20. [Clustering for Private Interest-based Advertising](https://dl.acm.org/doi/pdf/10.1145/3447548.3467180) |
| 99 | +Author(Institute): Alessandro Epasto(Google) |
| 100 | +KeyWords: Interest-based advertising; clustering; anonymity; privacy |
| 101 | +Dataset: Million Song; MovieLens |
| 102 | + |
| 103 | +21. [Diversity driven Query Rewriting in Search Advertising](https://arxiv.org/pdf/2106.03816.pdf) |
| 104 | +Author(Institute): Nikit Begwani(Microsoft二作) |
| 105 | +KeyWords: sponsored search; query rewriting; natural language generation |
| 106 | + |
| 107 | +22. [Exploration in Online Advertising Systems with Deep Uncertainty-Aware Learning](https://arxiv.org/pdf/2012.02298.pdf) |
| 108 | +Author(Institute): Chao Du(Alibaba) |
| 109 | +KeyWords: click-through rate (CTR); exploration-exploitation trade-off; advertising system; Gaussian process |
| 110 | +Dataset: Amazon |
| 111 | + |
| 112 | +23. [Neural Auction: End-to-End Learning of Auction Mechanisms for E-Commerce Advertising](https://arxiv.org/pdf/2106.03593.pdf) |
| 113 | +Author(Institute): Xiangyu Liu(Alibaba) |
| 114 | +KeyWords: Learning-based Mechanism Design; Neural Auction; E-commerce Advertising |
| 115 | +Dataset: Taobao |
| 116 | + |
| 117 | +24. [Reinforcing Pretrained Models for Generating Attractive Text Advertisements](https://www.microsoft.com/en-us/research/uploads/prod/2021/06/KDD2021_AdGen_camera_ready.pdf) |
| 118 | +Author(Institute): Xiting Wang(Microsoft) |
| 119 | +KeyWords: Advertisement Generation; Pretrained Language Models; Reinforcement Learning; Natural Language Generation |
| 120 | +Dataset: Microsoft Bing |
| 121 | + |
| 122 | +25. [Efficient Collaborative Filtering via Data Augmentation and Step-size Optimization](https://dl.acm.org/doi/abs/10.1145/3447548.3467380) |
| 123 | +Author(Institute): Xuejun Liao(SAS Institute Inc) |
| 124 | +KeyWords: Collaborative Filtering; Data Augmentation |
| 125 | +Dataset: MovieLens 1M |
| 126 | + |
| 127 | +26. [Efficient Data-specific Model Search for Collaborative Filtering](https://arxiv.org/pdf/2106.07453.pdf) |
| 128 | +Author(Institute): Quanming Yao(4Paradigm) |
| 129 | +KeyWords: Collaborative Filtering |
| 130 | +Dataset: MovieLens-100K; MovieLens-1M; Yelp; Amazon-Book |
| 131 | + |
| 132 | +27. [ML-based Visualization Recommendation: Learning to Recommend Visualizations from Data](https://arxiv.org/pdf/2009.12316.pdf) |
| 133 | +Author(Institute): Ryan A. [ Rossi(Adobe二作) |
| 134 | +KeyWords: Visualization recommendation; learning-based visualization recommendation; data visualization; machine learning; deep learning |
| 135 | +Dataset: Plot.ly |
| 136 | + |
| 137 | +28. [PURE: Positive-Unlabeled Recommendation with Generative Adversarial Network](https://jianpeng-xu.github.io/publications/2021-zhou-KDD-PURE.pdf) |
| 138 | +Author(Institute): Jianpeng Xu(Walmart) |
| 139 | +KeyWords: Recommender systems; Positive-unlabeled learning |
| 140 | +Dataset: Movielens; Yelp |
| 141 | + |
| 142 | +29. [Table2Charts: Recommending Charts by Learning Shared Table Representations](https://arxiv.org/pdf/2008.11015.pdf) |
| 143 | +Author(Institute): Mengyu Zhou(Microsoft) |
| 144 | +KeyWords: Table2seq; chart recommendation; deep Q-learning; copying mechanism; search sampling; transfer learning; table representations |
| 145 | +Dataset: Movielens; Yelp |
| 146 | + |
| 147 | +30. [Automated Loss Function Search in Recommendations](https://arxiv.org/pdf/2106.06713.pdf) |
| 148 | +Author(Institute): Chong Wang(Bytedance四作) |
| 149 | +KeyWords: AutoML; Recommender Systems; Loss Functions |
| 150 | +Dataset: Criteo; ML-20m |
| 151 | + |
| 152 | +31. [Bootstrapping Recommendations at Chrome Web Store](https://dl.acm.org/doi/pdf/10.1145/3447548.3467099) |
| 153 | +Author(Institute): Zhen Qin(Google) |
| 154 | +KeyWords: learning to rank; generalized additive models; text embedding |
| 155 | +Dataset: CWS |
| 156 | + |
| 157 | +32. [Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems](https://arxiv.org/pdf/2005.12964.pdf) |
| 158 | +Author(Institute): Chang Zhou(Alibaba) |
| 159 | +KeyWords: candidate generation; bias reduction; inverse propensity weighting; contrastive learning; negative sampling |
| 160 | +Dataset: ML-1M; Beauty; Steam |
| 161 | + |
| 162 | +33. [Device-Cloud Collaborative Learning for Recommendation](https://arxiv.org/pdf/2104.06624.pdf) |
| 163 | +Author(Institute): Jiangchao Yao(Alibaba) |
| 164 | +KeyWords: On-device Intelligence; Cloud Computing |
| 165 | +Dataset: Amazon; Movielens-1M; Taobao |
| 166 | + |
| 167 | +34. [FleetRec: Large-Scale Recommendation Inference on Hybrid GPU-FPGA Clusters](https://www.research-collection.ethz.ch/bitstream/handle/20.500.11850/485153/1/FleetRec_camera_ready.pdf) |
| 168 | +Author(Institute): Kai Zeng(Alibaba四作) |
| 169 | +KeyWords: scalable recommendation |
| 170 | + |
| 171 | +35. [Hierarchical Training: Scaling Deep Recommendation Models on Large CPU Clusters](https://dl.acm.org/doi/abs/10.1145/3447548.3467084) |
| 172 | +Author(Institute): Jiangchao Yao(Facebook) |
| 173 | +KeyWords: CPU Clusters |
| 174 | + |
| 175 | +36. [Leveraging Tripartite Interaction Information from Live Stream E-Commerce for Improving Product Recommendation](https://arxiv.org/pdf/2106.03415.pdf) |
| 176 | +Author(Institute): Zhuoxuan Jiang(Tencent二作); Dong-Dong Chen(JD三作); Dongsheng Li (Microsoft五作) |
| 177 | +KeyWords: graph representation learning; multi-task learning; live streaming E-Commence; product recommendation |
| 178 | +Dataset: LSEC-Small; LSECLarge |
| 179 | + |
| 180 | +37. [Sliding Spectrum Decomposition for Diversified Recommendation](https://arxiv.org/pdf/2107.05204.pdf) |
| 181 | +Author(Institute): Yanhua Huang(Xiaohongshu) |
| 182 | +KeyWords: Diversified Recommendation; Sliding Spectrum Decomposition; Item Embedding; Determinantal Point Process; CB2CF |
| 183 | + |
| 184 | +38. [Towards the D-Optimal Online Experiment Design for Recommender Selection](https://dl.acm.org/doi/abs/10.1145/3447548.3467192) |
| 185 | +Author(Institute): Da Xu(Walmart) |
| 186 | +KeyWords: Recommender Selection |
| 187 | +Dataset: Walmart |
0 commit comments