RENN

Code and Data for RENN published at ASE 2019

• Code and some pretrained models of the proposed DL technique are in folder rnns. We also include the implementation of the baseline models used in this paper. In particular, stats.py is used for some necessary statistics; ValueSet_RNN.py contains the implementation of each model and training evaluation; conditional_main.py and test.py are used for training and testing respectively.

• Code of crash analysis is in folder crash_analysis. RENN takes the memory regions predicted by deep learning and leverage the alias relationship to assist reverse execution.

• Code of Intel Pin tools to record ground truth is in folder pin_tools.