MGDCF: Distance Learning via Markov Graph Diffusion for Neural Collaborative Filtering

TL;DR

This paper introduces MGDCF, a novel approach that models GNN-based collaborative filtering as a Markov process, revealing their relation to traditional network embedding methods and emphasizing the importance of ranking losses.

Contribution

It presents MGDCF, a Markov graph diffusion model that generalizes GNN-based CF models and clarifies their connection to traditional network representation learning.

Findings

MGDCF achieves competitive performance without GNN training.

Ranking loss InfoBPR significantly improves CF results.

GNNs can be viewed as untrainable Markov processes for context construction.

Abstract

Graph Neural Networks (GNNs) have recently been utilized to build Collaborative Filtering (CF) models to predict user preferences based on historical user-item interactions. However, there is relatively little understanding of how GNN-based CF models relate to some traditional Network Representation Learning (NRL) approaches. In this paper, we show the equivalence between some state-of-the-art GNN-based CF models and a traditional 1-layer NRL model based on context encoding. Based on a Markov process that trades off two types of distances, we present Markov Graph Diffusion Collaborative Filtering (MGDCF) to generalize some state-of-the-art GNN-based CF models. Instead of considering the GNN as a trainable black box that propagates learnable user/item vertex embeddings, we treat GNNs as an untrainable Markov process that can construct constant context features of vertices for a traditional NRL model that encodes context features with a fully-connected layer. Such simplification can help us to better understand how GNNs benefit CF models. Especially, it helps us realize that ranking losses play crucial roles in GNN-based CF tasks. With our proposed simple yet powerful ranking loss InfoBPR, the NRL model can still perform well without the context features constructed by GNNs. We conduct experiments to perform detailed analysis on MGDCF.