Miltos Allamanis and Marc Brockschmidt take us through a new paper:
Finding and fixing bugs in code is a time-consuming, and often frustrating, part of everyday work for software developers. Can deep learning address this problem and help developers deliver better software, faster? In a new paper, Self-Supervised Bug Detection and Repair, presented at the 2021 Conference on Neural Information Processing Systems (NeurIPS 2021), we show a promising deep learning model, which we call BugLab. BugLab can be taught to detect and fix bugs, without using labelled data, through a “hide and seek” game.
I think there’s a lot more research required before we get to the point where this is useful in practical circumstances, but it’s exciting to see.