A Coverage-Guided Fuzzing Framework based on Genetic Algorithm for Neural Networks

Due to the inherent difference between neural network and traditional software, it is very difficult to test it. At present, the use of fuzzing methods may be an effective exploration direction. We choose coverage-guided fuzzing as a method to test neural networks, and use neuron coverage as a cover...

Full description

Saved in:

Bibliographic Details
Published in:	2021 8th International Conference on Dependable Systems and Their Applications (DSA) pp. 352 - 358
Main Authors:	Yi, Gaolei, Yang, Xiaoyu, Huang, Pu, Wang, Yichen
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-08-2021
Subjects:	Biological neural networks Fuzzing genetic algorithm Genetic algorithms Measurement neural network Neurons Software
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Due to the inherent difference between neural network and traditional software, it is very difficult to test it. At present, the use of fuzzing methods may be an effective exploration direction. We choose coverage-guided fuzzing as a method to test neural networks, and use neuron coverage as a coverage metric during execution. The effectiveness of neuron coverage will be demonstrated through experiments. On this basis, we designed a genetic algorithm-based fuzzing framework for neural networks, attempting to achieve greater coverage in a shorter time. And through the method of experimental comparison, the test efficiency of the framework is verified.
ISSN:	2767-6684
DOI:	10.1109/DSA52907.2021.00054