Application Specific Compression of Deep Learning Models
This addresses the need for more efficient and effective model deployment in specific applications, though it is incremental as it builds on existing compression techniques by adding application-specific customization.
The paper tackles the problem of compressing deep learning models without considering the target application, proposing Application Specific Compression (ASC) to prune redundant components for specific tasks, resulting in compressed models that outperform existing compression methods and off-the-shelf compressed models on BERT models for tasks like Extractive QA, Natural Language Inference, and Paraphrase Identification.
Large Deep Learning models are compressed and deployed for specific applications. However, current Deep Learning model compression methods do not utilize the information about the target application. As a result, the compressed models are application agnostic. Our goal is to customize the model compression process to create a compressed model that will perform better for the target application. Our method, Application Specific Compression (ASC), identifies and prunes components of the large Deep Learning model that are redundant specifically for the given target application. The intuition of our work is to prune the parts of the network that do not contribute significantly to updating the data representation for the given application. We have experimented with the BERT family of models for three applications: Extractive QA, Natural Language Inference, and Paraphrase Identification. We observe that customized compressed models created using ASC method perform better than existing model compression methods and off-the-shelf compressed models.