No’a: A Neural Network approach to automated exploitation
The need for automated software auditing in the recent years has sparked the creation of fuzzing. Fuzzing is a software auditing method that attempts to find a crash by generating enormous amounts of input and running each against a program until one crashes the program. These crashes are indications of a potential exploit. Though fuzzing is the closest thing we have to automated exploit discovery, the time to discovering a crash is prohibitively long. No’a is a neural network approach to increasing the speed of fuzzing through recursive pattern recognition in exploit generation.