Statistical Model Compression for Small-Footprint Natural Language Understanding
arXiv:1807.07520
Abstract
In this paper we investigate statistical model compression applied to natural language understanding (NLU) models. Small-footprint NLU models are important for enabling offline systems on hardware restricted devices, and for decreasing on-demand model loading latency in cloud-based systems. To compress NLU models, we present two main techniques, parameter quantization and perfect feature hashing. These techniques are complementary to existing model pruning strategies such as L1 regularization. We performed experiments on a large scale NLU system. The results show that our approach achieves 14-fold reduction in memory usage compared to the original models with minimal predictive performance impact.
arXiv commentInterspeech 2018
References in corpus (4)
- Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
- Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
- Experiments with Random Projection
- Just ASK: Building an Architecture for Extensible Self-Service Spoken Language Understanding