Convolutional Gaussian Embeddings for Personalized Recommendation with Uncertainty

TL;DR

This paper introduces a deep recommendation framework using Gaussian embeddings to better model user uncertainty, employing Monte-Carlo sampling and CNNs for improved personalized recommendations, validated by experiments on benchmark datasets.

Contribution

It presents a novel Gaussian embedding approach for recommender systems that effectively captures user uncertainty, enhancing recommendation accuracy over existing models.

Findings

Gaussian embeddings better capture user uncertainty

Proposed method outperforms state-of-the-art models

Framework effectively utilizes Monte-Carlo sampling and CNNs

Abstract

Most of existing embedding based recommendation models use embeddings (vectors) corresponding to a single fixed point in low-dimensional space, to represent users and items. Such embeddings fail to precisely represent the users/items with uncertainty often observed in recommender systems. Addressing this problem, we propose a unified deep recommendation framework employing Gaussian embeddings, which are proven adaptive to uncertain preferences exhibited by some users, resulting in better user representations and recommendation performance. Furthermore, our framework adopts Monte-Carlo sampling and convolutional neural networks to compute the correlation between the objective user and the candidate item, based on which precise recommendations are achieved. Our extensive experiments on two benchmark datasets not only justify that our proposed Gaussian embeddings capture the uncertainty of users very well, but also demonstrate its superior performance over the state-of-the-art recommendation models.