Deep Program Structure Modeling Through Multi-Relational Graph-based Learning
Guixin Ye, Zhanyong Tang, Huanting Wang, Dingyi Fang, Jianbin Fang, Songfang Huang, Zheng Wang
Abstract
Deep learning is emerging as a promising technique for building predictive models to support code-related tasks like performance optimization and code vulnerability detection. One of the critical aspects of building a successful predictive model is having the right representation to characterize the model input for the given task. Existing approaches in the area typically treat the program structure as a sequential sequence but fail to capitalize on the rich semantics of data and control flow information, for which graphs are a proven representation structure.
Topics & Concepts
Computer scienceControl flow graphTask (project management)Code (set theory)Representation (politics)Artificial intelligenceDeep learningGraphSemantics (computer science)Data structureExternal Data RepresentationMachine learningVulnerability (computing)Theoretical computer scienceProgram optimizationData modelingProgramming languageSoftware engineeringCompilerLawPolitical sciencePoliticsComputer securityEconomicsSet (abstract data type)ManagementSoftware Engineering ResearchSoftware System Performance and ReliabilitySoftware Testing and Debugging Techniques