Research Project Title:
Modeling Concurrency Bugs in Software Using Machine Learning
abstract:As Artificial Intelligence has started to gain a lot of traction in the last few years, we believe one rising problem where novel machine learning techniques could be applied is the increasingly complex issue of ensuring software safety and security. Software would be passed into a machine learning model which would then make a prediction on where in the code bugs are more likely to occur. This sort of automation would speed up the process of debugging, since the programmer could spend more time fixing the bug and less time looking for it. On the other hand, it could alert the programmer of sections in the code that were previously thought to be safe, thus reducing the chance for unexpected behavior of software.
"I am participating in this SuperUROP project as I think it provides an amazing learning opportunity within the emerging field of AI, and if successful, this project would hugely benefit software engineering. Having taken machine Learning courses previously and having just finished a 20-week internship in data science, I am thrilled to see how I can use my experience to drive this project forward."