Handling Large-scale Cardinality in building recommendation systems
This addresses a scalability issue for companies like Uber building large-scale recommendation systems, though it appears incremental as it builds on existing techniques.
The paper tackles the problem of model degradation and increased model size caused by high-cardinality UUIDs in recommendation systems by proposing a bag-of-words approach with layer sharing, resulting in reduced model size and improved performance as shown in Uber experiments.
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.