Latent Structured Ranking

TL;DR

This paper introduces a scalable latent structured ranking method that improves the quality of ranked lists by considering their overall structure, leading to better diversity and relevance in recommendation and retrieval tasks.

Contribution

It proposes a novel scalable approach for learning latent structured rankings that enhances the coherence and diversity of top predictions in large-scale applications.

Findings

Improves ranking quality in image annotation tasks

Enhances music recommendation accuracy

Outperforms existing methods in empirical evaluations

Abstract

Many latent (factorized) models have been proposed for recommendation tasks like collaborative filtering and for ranking tasks like document or image retrieval and annotation. Common to all those methods is that during inference the items are scored independently by their similarity to the query in the latent embedding space. The structure of the ranked list (i.e. considering the set of items returned as a whole) is not taken into account. This can be a problem because the set of top predictions can be either too diverse (contain results that contradict each other) or are not diverse enough. In this paper we introduce a method for learning latent structured rankings that improves over existing methods by providing the right blend of predictions at the top of the ranked list. Particular emphasis is put on making this method scalable. Empirical results on large scale image annotation and music recommendation tasks show improvements over existing approaches.