Papers › UltraGCN: Ultra Simplification of Graph Convolutional Networks for Recommendation

UltraGCN: Ultra Simplification of Graph Convolutional Networks for Recommendation

28 Oct 2021arXiv:2110.15114archive 2025-07-28

Kelong Mao, Jieming Zhu, Xi Xiao, Biao Lu, Zhaowei Wang, Xiuqiang He

With the recent success of graph convolutional networks (GCNs), they have been widely applied for recommendation, and achieved impressive performance gains. The core of GCNs lies in its message passing mechanism to aggregate neighborhood information. However, we observed that message passing largely slows down the convergence of GCNs during training, especially for large-scale recommender systems, which hinders their wide adoption. LightGCN makes an early attempt to simplify GCNs for collaborative filtering by omitting feature transformations and nonlinear activations. In this paper, we take one step further to propose an ultra-simplified formulation of GCNs (dubbed UltraGCN), which skips infinite layers of message passing for efficient recommendation. Instead of explicit message passing, UltraGCN resorts to directly approximate the limit of infinite-layer graph convolutions via a constraint loss. Meanwhile, UltraGCN allows for more appropriate edge weight assignments and flexible adjustment of the relative importances among different types of relationships. This finally yields a simple yet effective UltraGCN model, which is easy to implement and efficient to train. Experimental results on four benchmark datasets show that UltraGCN not only outperforms the state-of-the-art GCN models but also achieves more than 10x speedup over LightGCN. Our source code will be available at https://reczoo.github.io/UltraGCN.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

reczoo/RecZoo officialpytorch report
shuyao-wang/dsl mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Collaborative FilteringRecommendation Systems

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Collaborative Filtering Gowalla UltraGCN NDCG@20 0.1580 #7 of 11 Archive leaderboard report
Collaborative Filtering Gowalla UltraGCN Recall@20 0.1862 #7 of 11 Archive leaderboard report
Collaborative Filtering Gowalla Emb-GCN Recall@20 0.1862 #8 of 11 Archive leaderboard report
Collaborative Filtering MovieLens 1M UltraGCN NDCG@20 0.2642 #2 of 4 Archive leaderboard report
Collaborative Filtering MovieLens 1M UltraGCN Recall@20 0.2787 #2 of 4 Archive leaderboard report
Collaborative Filtering Yelp2018 UltraGCN NDCG@20 0.0561 #6 of 9 Archive leaderboard report
Collaborative Filtering Yelp2018 UltraGCN Recall@20 0.0683 #6 of 9 Archive leaderboard report
Recommendation Systems Amazon-Book Emb-GCN Recall@20 0.0681 #7 of 16 Archive leaderboard report
Recommendation Systems Amazon-Book Emb-GCN nDCG@20 0.0556 #7 of 16 Archive leaderboard report
Recommendation Systems Gowalla Emb-GCN Recall@20 0.1862 #6 of 13 Archive leaderboard report
Recommendation Systems Gowalla Emb-GCN nDCG@20 0.1580 #6 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

GCNLightGCN

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections