第36届神经信息处理系统会议NeurlPS2022将于11月28日至12月9日举行。大会将持续举行两周，第一周将在美国新奥尔良举行现场会议，第二周改为线上会议。根据官网邮件中给出的数据，本届会议共有10411篇论文投稿，接收率为25.6％。
官网链接：
https://nips.cc/Conferences/2022/Schedule?type=Poster
本文从上述官网链接接收列表中筛选出与推荐系统、数据偏差以及图相关的论文供大家学习，其中与推荐系统有关的论文16篇。本次论文整理涉及到众多推荐系统领域的子方向，比如推荐系统中的嵌入表存放问题、算法选择问题、推荐系统中具有点击后信息的广义延迟反馈模型、用于训练推荐模型的缓存增强batch内重要性重采样问题、基于图卷积网络的推荐系统、推荐系统中的注入攻击分析、多样性推荐、基于自监督学习的推荐系统、大规模多用途的推荐系统数据集等。
其中部分论文已上传到Arxiv，大家可以自行下载进行阅读，也可以前往每周的论文周报进行查看。
DreamShard: Generalizable Embedding Table Placement for Recommender Systems | ||||||
|---|---|---|---|---|---|---|
RecZilla: Algorithm Selection for Recommender Systems | ||||||
APG: Adaptive Parameter Generation Network for Click-Through Rate Prediction | ||||||
Generalized Delayed Feedback Model with Post-Click Information in Recommender Systems | ||||||
Cache-Augmented Inbatch Importance Resampling for Training Recommender Retriever | ||||||
Graph Convolution Network based Recommender Systems: Learning Guarantee and Item Mixture Powered Strategy | ||||||
Infinite Recommendation Networks: A Data-Centric Approach | ||||||
Revisiting Injective Attacks on Recommender Systems | ||||||
Diversified Recommendations for Agents with Adaptive Preferences | ||||||
Debiased, Longitudinal and Coordinated Drug Recommendation through Multi-Visit Clinic Records | ||||||
GBA: A Tuning-free Approach to Switch between Synchronous and Asynchronous Training for Recommendation Models | ||||||
Recommender Forest for Efficient Retrieval | ||||||
The trade-offs of model size in large recommendation models : A 10000 × compressed criteo-tb DLRM model (100 GB parameters to mere 10MB) | ||||||
Contrastive Graph Structure Learning via Information Bottleneck for Recommendation | ||||||
Tenrec: A Large-scale Multipurpose Benchmark Dataset for Recommender Systems | ||||||
Incorporating Bias-aware Margins into Contrastive Loss for Collaborative Filtering |
