Sign In

Communications of the ACM

ACM Careers

Intel Opens ControlFlag Machine Learning System for Improving Code Quality

magnifying glass on bug in binary code, illustration

Credit: TechGig

Intel has made its ControlFlag software, which uses machine learning to improve code quality, open source and accessible on Github. The research project toolkit allows, based on a model trained on a large amount of existing code, to identify various errors and anomalies in source texts written in high-level languages such as C / C ++. The system detects various kinds of problems, from typos and incorrect combinations of types, to missing checks for null values in pointers and problems with memory. 

In a few analyses of source code repositories, ControlFlag found hundreds of confirmed software defects in proprietary, production-quality software, according to a post by Justin Gottschlich, principal AI scientist at Intel Labs.

The system self-learns by building a statistical model of the existing array of open source code published on GitHub and similar public repositories.

From It's FOSS
View Full Article


No entries found