Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

10-2022

Abstract

Although large pre-trained models of code have delivered significant advancements in various code processing tasks, there is an impediment to the wide and fluent adoption of these powerful models in software developers’ daily workflow: these large models consume hundreds of megabytes of memory and run slowly on personal devices, which causes problems in model deployment and greatly degrades the user experience. It motivates us to propose Compressor, a novel approach that can compress the pre-trained models of code into extremely small models with negligible performance sacrifice. Our proposed method formulates the design of tiny models as simplifying the pre-trained model architecture: searching for a significantly smaller model that follows an architectural design similar to the original pre-trained model. Compressor proposes a genetic algorithm (GA)-based strategy to guide the simplification process. Prior studies found that a model with higher computational cost tends to be more powerful. Inspired by this insight, the GA algorithm is designed to maximize a model’s Giga floating-point operations (GFLOPs), an indicator of the model computational cost, to satisfy the constraint of the target model size. Then, we use the knowledge distillation technique to train the small model: unlabelled data is fed into the large model and the outputs are used as labels to train the small model. We evaluate Compressor with two state-of-the-art pre-trained models, i.e., CodeBERT and GraphCodeBERT, on two important tasks, i.e., vulnerability prediction and clone detection. We use our method to compress pre-trained models to a size (3 MB), which is 160× smaller than the original size. The results show that compressed CodeBERT and GraphCodeBERT are 4.31× and 4.15× faster than the original model at inference, respectively. More importantly, they maintain 96.15% and 97.74% of the original performance on the vulnerability prediction task. They even maintain higher ratios (99.20% and 97.52%) of the original performance on the clone detection task

Keywords

Model compression, Genetic algorithm, Pre-trained models

Discipline

Software Engineering

Research Areas

Software and Cyber-Physical Systems

Publication

ASE '22: Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering, Rochester, MI, October 10-14

First Page

1

Last Page

12

ISBN

9781450394758

Identifier

10.1145/3551349.3556964

Publisher

ACM

City or Country

New York

Copyright Owner and License

Authors

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Additional URL

https://doi.org/10.1145/3551349.3556964

Share

COinS