Fully Hyperbolic Graph Convolution Network for Recommendation

TL;DR

This paper introduces a fully hyperbolic graph convolution network for recommendation systems, leveraging hyperbolic space to better capture hierarchical user-item relationships and outperform Euclidean models.

Contribution

It proposes a novel hyperbolic GCN model where all operations are in hyperbolic space, improving embedding efficiency and accuracy over traditional Euclidean approaches.

Findings

Outperforms Euclidean and hyperbolic baselines on benchmark datasets

Achieves comparable performance with lower embedding dimensions

Better captures hierarchical user-item relationships

Abstract

Recently, Graph Convolution Network (GCN) based methods have achieved outstanding performance for recommendation. These methods embed users and items in Euclidean space, and perform graph convolution on user-item interaction graphs. However, real-world datasets usually exhibit tree-like hierarchical structures, which make Euclidean space less effective in capturing user-item relationship. In contrast, hyperbolic space, as a continuous analogue of a tree-graph, provides a promising alternative. In this paper, we propose a fully hyperbolic GCN model for recommendation, where all operations are performed in hyperbolic space. Utilizing the advantage of hyperbolic space, our method is able to embed users/items with less distortion and capture user-item interaction relationship more accurately. Extensive experiments on public benchmark datasets show that our method outperforms both Euclidean and hyperbolic counterparts and requires far lower embedding dimensionality to achieve comparable performance.