论文索引
阅读算法的原始论文是学习它的最好途径,为什么呢?一是网上的博客资料质量良莠不齐,有时可能是错的;二是论文中一定会重点阐述算法的核心创新点到底是怎么来的,以及这个算法的完整上下文。
集成学习
- CatBoost: gradient boosting with categorical features support:catboost的全面介绍
- CatBoost: unbiased boosting with categorical features:详细论述如何处理类别特征以及标签泄露问题
图算法
社区发现
- Community structure in social and biological networks: GN算法,社区发现领域的第一篇论文
- Finding and evaluating community structure in networks: 详细介绍模块度函数(modurity), IV部分
- Fast unfolding of communities in large networks: state-of-art的算法Louvain
- Near linear time algorithm to detect community structures in large-scale networks
Towards Real-Time Community Detection in Large Networks : 标签传播算法 - Uncovering the overlapping community structure of complex networks in nature and society: 第一篇重叠社区发现算法
- Detecting the overlapping and hierarchical community structure in complex networks: 基于局部扩展的算法
图嵌入
- DeepWalk: Online Learning of Social Representations
- LINE: Large-scale Information Network Embedding
- node2vec: Scalable Feature Learning for Networks
图神经网络
- Semi-Supervised Classification With Graph Convolution Networks :半监督
- Graph Attention Networks
- Inductive Representation Learning on Large Graphs :Graphsage
- Graph Convolutional Neural Networks for Web-ScaleRecommender Systems :Pinsage
风控算法
- Isolation Forest
- Isolation-based Anomaly Detection
- Outlier Detection with Autoencoder Ensembles
- CopyCatch: Stopping Group Attacks by Spotting LockstepBehavior in Social Networks : 基于用户行为和时间的双聚类算法
- Uncovering Large Groups of Active Malicious Accounts inOnline Social Networks :Synchrotrap
- FRAUDAR: Bounding Graph Fraud in the Face ofCamouflage
计算广告
点击率优化
- Practical Lessons from Predicting Clicks on Ads at Facebook: GBDT+LR
- Wide & Deep Learning for Recommender Systems: lr memorization & deep generalization
- Factorization Machines: fm特征组合
- DeepFM: A Factorization-Machine based Neural Network for CTR Prediction: fm替换人工特征组合
- Field-aware Factorization Machines for CTR Prediction
- Deep Interest Network for Click-Through Rate Prediction: 阿里巴巴din
评论系统未开启,无法评论!