AM-GCN: Adaptive Multi-channel Graph Convolutional Networks 文中提出在復(fù)雜網(wǎng)絡(luò)中,現(xiàn)有的GCN方法融合節(jié)點(diǎn)屬性信息...
AM-GCN: Adaptive Multi-channel Graph Convolutional Networks 文中提出在復(fù)雜網(wǎng)絡(luò)中,現(xiàn)有的GCN方法融合節(jié)點(diǎn)屬性信息...
Deep Anomaly Detection on Attributed Networks 提出了DOMINANT(Deep Anomaly Detection on Att...
Time Series Shapelets: A New Primitive for Data Mining 對(duì)于時(shí)序數(shù)據(jù),提出了一種新的特征,名為shapelet。shap...
Dynamic Heterogeneous Graph Neural Network for Real-time Event Prediction 本文是滴滴發(fā)在KDD202...
GPT-GNN: Generative Pre-Training of Graph Neural Networks 文中指出訓(xùn)練GNN需要大量和任務(wù)對(duì)應(yīng)的標(biāo)注數(shù)據(jù),這在很多時(shí)...
Deep Neural Networks for YouTube Recommendations 文中把整個(gè)推薦過(guò)程分成兩個(gè)步驟:?deep candidate genera...
xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems 目...
Feature Generation by Convolutional Neural Network for Click-Through Rate Prediction 目標(biāo)...
A Convolutional Click Prediction Model 目標(biāo):CTR預(yù)估 Convolution Layer 對(duì)于有n個(gè)元素的輸入,先做embeddin...
Deep Interest Evolution Network for Click-Through Rate Prediction 目標(biāo):CTR預(yù)估文中指出以前的模型忽略了用...
Deep Interest Network for Click-Through Rate Prediction 本文指出,在基于embedding和MLP的模型中,用戶的各種...
Attentional Factorization Machines: Learning the Weight of Feature Interactions via Att...
Neural Factorization Machines for Sparse Predictive Analytics 文中提到了對(duì)于稀疏特征交互的處理。FM以線性的方式...
DeepFM: A Factorization-Machine based Neural Network for CTR Prediction 目標(biāo):CTR預(yù)估文中指出以前的...
Deep Learning over Multi-field Categorical Data – A Case Study on User 目標(biāo):CTR預(yù)估。本文提出了FN...
Deep & Cross Network for Ad Click Predictions 目標(biāo):CTR預(yù)估。本文提出了Deep & Cross Network(DCN)能夠...
Wide & Deep Learning for Recommender Systems 文中指出推薦系統(tǒng)的一個(gè)挑戰(zhàn)是同時(shí)實(shí)現(xiàn)memorization和generalizat...
Product-based Neural Networks for User Response Prediction 目標(biāo):CTR預(yù)估。文中指出LR,GBDT,F(xiàn)M雖然在工業(yè)...
Neural Collaborative Filtering 文中指出雖然之前有一些工作使用deep learning來(lái)解決推薦問(wèn)題,但這些工作基本上使用deep learn...
Deep Crossing: Web-Scale Modeling without Manually Crafted Combinatorial Features 文中開篇講...