M^2VAE: Multi-Modal Multi-View Variational Autoencoder for Cold-start Item Recommendation

TL;DR

M^2VAE is a novel generative model that leverages multi-modal and multi-view features to improve cold-start item recommendation by disentangling shared and unique information and modeling user preferences adaptively.

Contribution

It introduces a multi-view variational autoencoder that explicitly models common and modality-specific features, incorporating contrastive learning and a preference-guided mixture of experts for enhanced cold-start recommendations.

Findings

Outperforms existing methods on real-world datasets.

Effectively disentangles shared and unique feature representations.

Eliminates the need for pretraining through contrastive learning.

Abstract

Cold-start item recommendation is a significant challenge in recommendation systems, particularly when new items are introduced without any historical interaction data. While existing methods leverage multi-modal content to alleviate the cold-start issue, they often neglect the inherent multi-view structure of modalities, the distinction between shared and modality-specific features. In this paper, we propose Multi-Modal Multi-View Variational AutoEncoder (M^2VAE), a generative model that addresses the challenges of modeling common and unique views in attribute and multi-modal features, as well as user preferences over single-typed item features. Specifically, we generate type-specific latent variables for item IDs, categorical attributes, and image features, and use Product-of-Experts (PoE) to derive a common representation. A disentangled contrastive loss decouples the common view from unique views while preserving feature informativeness. To model user inclinations, we employ a preference-guided Mixture-of-Experts (MoE) to adaptively fuse representations. We further incorporate co-occurrence signals via contrastive learning, eliminating the need for pretraining. Extensive experiments on real-world datasets validate the effectiveness of our approach.