Handling Large-scale Cardinality in building recommendation systems

TL;DR

This paper introduces two novel techniques combining bag-of-words and layer sharing to effectively handle high-cardinality UUID features in recommendation systems, reducing model size and improving performance.

Contribution

The paper presents innovative methods to address high cardinality in recommendation systems, significantly reducing model size and enhancing performance compared to existing approaches.

Findings

Reduced model size by leveraging layer sharing.

Improved recommendation accuracy in Uber use cases.

Validated effectiveness through offline and online experiments.

Abstract

Effective recommendation systems rely on capturing user preferences, often requiring incorporating numerous features such as universally unique identifiers (UUIDs) of entities. However, the exceptionally high cardinality of UUIDs poses a significant challenge in terms of model degradation and increased model size due to sparsity. This paper presents two innovative techniques to address the challenge of high cardinality in recommendation systems. Specifically, we propose a bag-of-words approach, combined with layer sharing, to substantially decrease the model size while improving performance. Our techniques were evaluated through offline and online experiments on Uber use cases, resulting in promising results demonstrating our approach's effectiveness in optimizing recommendation systems and enhancing their overall performance.