Lessons Learned on Machine Learning for Computer Security

Daniel Arp, Erwin Quiring, Feargus Pendlebury, Alexander Warnecke, Fabio Pierazzi, Christian Wressnegger, Lorenzo Cavallaro, Konrad Rieck

Abstract: We identify 10 generic pitfalls that can affect the experimental outcome of AI driven solutions in computer security. We find that they are prevalent in the literature and provide recommendations for overcoming them in the future.