学习并整理一些经典的推荐系统算法
目标：主要根据用户部分特征，从海量的物品库里，快速找回一小部分用户潜在感兴趣的物品，然后交给排序环节，
论文地址：word2vec Parameter Learning Explained
论文地址：DeepWalk: Online Learning of Social Representations
论文地址：LINE: Large-scale Information Network Embedding
论文地址：node2vec: Scalable Feature Learning for Networks
论文地址：Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba
论文地址：Deep Neural Networks for YouTube Recommendations
论文地址：Sampling-bias-corrected neural modeling for large corpus item recommendations
目标：精排环节可以融入较多特征，使用复杂模型，来精准地做个性化推荐。
论文地址：Collaborative Filtering Recommender Systems
论文地址：Matrix Factorization Techniques For Recommender Systems
论文地址：Predicting Clicks: Estimating the Click-Through Rate for New Ads
论文地址：Practical Lessons from Predicting Clicks on Ads at Facebook
论文地址：AutoRec: Autoencoders Meet Collaborative Filtering
论文地址：Deep & Cross Network for Ad Click Predictions
论文地址：Neural Collaborative Filtering
论文地址：Product-based Neural Networks for User Response Prediction
论文地址：Wide & Deep Learning for Recommender Systems
论文地址：DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
官方代码：DeepFM
论文地址：xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
官方代码：xDeepFM
论文地址：Deep Interest Network for Click-Through Rate Prediction
论文地址：Deep Interest Evolution Network for Click-Through Rate Prediction
目标：多目标排序通常是指有两个或两个以上的目标函数，目的是寻求一种排序使得所有的目标函数都达到最优或满意。 在工业界推荐系统中，大多是基于隐式反馈来进行推荐的，用户对推荐结果的满意度通常依赖很多指标（比如，淘宝基于点击，浏览深度（停留时间），加购，收藏，购买，重复购买，好评等行为的相关指标），在不同的推荐系统、不同时期、不同的产品形态下，这些指标的重要程度或者所代表的意义也会有所不同，如何优化最终推荐列表的顺序来使得众多指标在不同的场景下近可能达到最优或者满意，这就是一个多目标排序问题。
论文地址：An Overview of Multi-Task Learning in Deep Neural Networks
论文地址：Entire Space Multi-Task Model: An Effective Approach for Estimating Post-Click Conversion Rate
论文地址：Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts
论文地址：Recommending what video to watch next: a multitask ranking system
待更新......