UltraGCN: Ultra Simplification of Graph Convolutional Networks for Recommendation

TL;DR

UltraGCN simplifies graph convolutional networks for recommendation by avoiding message passing, leading to faster training and improved performance over existing GCN models, especially on large-scale datasets.

Contribution

Proposes UltraGCN, an ultra-simplified GCN model that approximates infinite-layer convolutions through a constraint loss, enhancing efficiency and flexibility.

Findings

Outperforms state-of-the-art GCN models on four benchmark datasets.

Achieves over 10x speedup compared to LightGCN.

Effective in large-scale recommender systems.

Abstract

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.