Towards Automated Software Testing with Generative Adversarial Networks
Xiujing Guo
Abstract
This paper discusses software testing methods based on Generative Adversarial Network (GAN). GAN is a generative model that can create new data instances that resemble training data. A GAN consists of a generator network and a discriminator network. In our testing scheme, the trained generator network is used as a test case generator. In addition, we propose a framework with GAN, which is a testing strategy used to increase the test coverage.
Topics & Concepts
DiscriminatorComputer scienceGenerator (circuit theory)Generative adversarial networkGenerative grammarAdversarial systemSoftwareScheme (mathematics)Software testingTest dataArtificial intelligenceMachine learningComputer engineeringSoftware engineeringDeep learningProgramming languageTelecommunicationsMathematical analysisPower (physics)PhysicsDetectorQuantum mechanicsMathematicsSoftware Testing and Debugging TechniquesAdversarial Robustness in Machine LearningAnomaly Detection Techniques and Applications