A Generic Network Compression Framework for Sequential Recommender Systems

Abstract

Sequential recommender systems (SRS) have become the key technology in capturing user's dynamic interests and generating high-quality recommendations. Current state-of-the-art sequential recommender models are typically based on a sandwich-structured deep neural network, where one or more middle (hidden) layers are placed between the input embedding layer and output softmax layer. In general, these models require a large number of parameters (such as using a large embedding dimension or a deep network architecture) to obtain their optimal performance. Despite the effectiveness, at some point, further increasing model size may be harder for model deployment in resource-constraint devices, resulting in longer responding time and larger memory footprint. To resolve the issues, we propose a compressed sequential recommendation framework, termed as CpRec, where two generic model shrinking techniques are employed. Specifically, we first propose a block-wise adaptive decomposition to approximate the input and softmax matrices by exploiting the fact that items in SRS obey a long-tailed distribution. To reduce the parameters of the middle layers, we introduce three layer-wise parameter sharing schemes. We instantiate CpRec using deep convolutional neural network with dilated kernels given consideration to both recommendation accuracy and efficiency. By the extensive ablation studies, we demonstrate that the proposed CpRec can achieve up to 4$\sim$8 times compression rates in real-world SRS datasets. Meanwhile, CpRec is faster during training\inference, and in most cases outperforms its uncompressed counterpart.

Accepted by SIGIR2020

References in corpus (11)

• Wide Residual Networks
• Compressing Deep Convolutional Networks using Vector Quantization
• Session-based Recommendations with Recurrent Neural Networks
• Adversarial Personalized Ranking for Recommendation
• Head/tail Breaks: A New Classification Scheme for Data with a Heavy-tailed Distribution
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
• Adaptive Input Representations for Neural Language Modeling
• Deep Neural Network Compression with Single and Multiple Level Quantization
• ShaResNet: reducing residual network parameter number by sharing weights
• CmnRec: Sequential Recommendations with Chunk-accelerated Memory Network
• GroupReduce: Block-Wise Low-Rank Approximation for Neural Language Model Shrinking

Cited by in corpus (3)

• StackRec: Efficient Training of Very Deep Sequential Recommender Models by Iterative Stacking
• Parameter-Efficient Transfer from Sequential Behaviors for User Modeling and Recommendation
• User-specific Adaptive Fine-tuning for Cross-domain Recommendations