Probabilistic Metric Learning with Adaptive Margin for Top-K Recommendation

TL;DR

This paper introduces a probabilistic metric learning model with adaptive margins for Top-K recommendation, leveraging Wasserstein distance and Gaussian distributions to better capture uncertainties and preferences, outperforming existing methods.

Contribution

The paper proposes a novel distance-based recommendation model using Gaussian distributions, adaptive margin generation, and Wasserstein distance to improve accuracy in Top-K recommendations.

Findings

Outperforms state-of-the-art models by 4-22% in recall@K

Utilizes Wasserstein distance to obey triangle inequality

Incorporates uncertainty modeling with Gaussian distributions

Abstract

Personalized recommender systems are playing an increasingly important role as more content and services become available and users struggle to identify what might interest them. Although matrix factorization and deep learning based methods have proved effective in user preference modeling, they violate the triangle inequality and fail to capture fine-grained preference information. To tackle this, we develop a distance-based recommendation model with several novel aspects: (i) each user and item are parameterized by Gaussian distributions to capture the learning uncertainties; (ii) an adaptive margin generation scheme is proposed to generate the margins regarding different training triplets; (iii) explicit user-user/item-item similarity modeling is incorporated in the objective function. The Wasserstein distance is employed to determine preferences because it obeys the triangle inequality and can measure the distance between probabilistic distributions. Via a comparison using five real-world datasets with state-of-the-art methods, the proposed model outperforms the best existing models by 4-22% in terms of recall@K on Top-K recommendation.