AI Paper Atlas

Research map / Graph learning / GNN architectures

Graph Convolution Network based Recommender Systems: Learning Guarantee and Item Mixture Powered Strategy

Leyan Deng, Defu Lian, Chenwang Wu, Enhong Chen

NeurIPS 2022 · Graph learning › GNN architectures · approach: Deep graph convolution and simplification · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Item Mixture (IMix) augmentationNGCF 3 layer · GowallaRecall@20: 0.1303 -> 0.1496Improved
add IMixNGCF 1 layer · Yelp2018Recall@20: 0.0419 -> 0.0492Improved
add IMixNGCF 1 layer · GowallaRecall@20: 0.1308 -> 0.1451Improved
add IMixLightGCN 3 layer · GowallaRecall@20: 0.1759 -> 0.1793Improved
add IMixLightGCN 2 layer · Yelp2018Recall@20: 0.0564 -> 0.0621Improved
add IMixLightGCN 3 layer · Yelp2018NDCG@20: 0.0492 -> 0.0527Improved
use IMix on weaker NGCF versus stronger LightGCNIMix · Gowalla and Yelp2018relative improvementMixed
train LightGCN with IMix beyond 50 epochsLightGCN-IMix · Gowalla and Yelp2018test Recall/NDCG over epochsImproved
increase number of GCN layers L (ablation)NGCF · Gowallageneralization gap (Recall/NDCG@K)Hurt
increase spectral norm of weight matrices (ablation)NGCF · Gowallageneralization gapHurt
use normalized graph (D^-1 A + I or D^-1/2 A D^-1/2 + I) instead of unnormalized A + I (ablation)NGCF · Gowallageneralization (Recall@20, NDCG@20)Improved
swap non-linear activation (LeakyReLU, ReLU, Sigmoid, Tanh, none) (ablation)NGCF · GowallaRecall@20, NDCG@20No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy