Hierarchical Latent Relation Modeling for Collaborative Metric Learning

TL;DR

This paper introduces a hierarchical collaborative metric learning model that captures complex user-item and item-item relations using memory-based attention, improving recommendation accuracy on real-world datasets.

Contribution

It proposes a novel hierarchical CML model that jointly models user-item and item-item relations, addressing limitations of fixed representations and relation heterogeneity.

Findings

Outperforms existing CML models on multiple datasets

Highlights the importance of joint relational modeling in recommendations

Identifies challenges of CML on very sparse data

Abstract

Collaborative Metric Learning (CML) recently emerged as a powerful paradigm for recommendation based on implicit feedback collaborative filtering. However, standard CML methods learn fixed user and item representations, which fails to capture the complex interests of users. Existing extensions of CML also either ignore the heterogeneity of user-item relations, i.e. that a user can simultaneously like very different items, or the latent item-item relations, i.e. that a user's preference for an item depends, not only on its intrinsic characteristics, but also on items they previously interacted with. In this paper, we present a hierarchical CML model that jointly captures latent user-item and item-item relations from implicit data. Our approach is inspired by translation mechanisms from knowledge graph embedding and leverages memory-based attention networks. We empirically show the relevance of this joint relational modeling, by outperforming existing CML models on recommendation tasks on several real-world datasets. Our experiments also emphasize the limits of current CML relational models on very sparse datasets.