Model Size Reduction Using Frequency Based Double Hashing for Recommender Systems

TL;DR

This paper introduces a hybrid hashing technique combining frequency hashing and double hashing to significantly reduce the size of recommender system models by around 90% without losing performance.

Contribution

It presents a novel hybrid hashing method that effectively compresses large DNN models in recommender systems while maintaining their accuracy.

Findings

Model size reduced by approximately 90%.

Performance remains comparable to baseline models.

Effective on multiple product surfaces.

Abstract

Deep Neural Networks (DNNs) with sparse input features have been widely used in recommender systems in industry. These models have large memory requirements and need a huge amount of training data. The large model size usually entails a cost, in the range of millions of dollars, for storage and communication with the inference services. In this paper, we propose a hybrid hashing method to combine frequency hashing and double hashing techniques for model size reduction, without compromising performance. We evaluate the proposed models on two product surfaces. In both cases, experiment results demonstrated that we can reduce the model size by around 90 % while keeping the performance on par with the original baselines.