{"id":1144,"date":"2017-12-08T10:23:15","date_gmt":"2017-12-08T15:23:15","guid":{"rendered":"http:\/\/www.jsylvest.com\/blog\/?p=1144"},"modified":"2017-12-08T10:23:15","modified_gmt":"2017-12-08T15:23:15","slug":"malconv","status":"publish","type":"post","link":"https:\/\/www.jsylvest.com\/blog\/2017\/12\/malconv\/","title":{"rendered":"MalConv: Lessons learned from Deep Learning on executables"},"content":{"rendered":"

I don't usually write up my technical work here, mostly because I spend enough hours as is doing technical writing. But a co-author, Jon Barker, recently wrote\u00a0a post on the NVIDIA Parallel For All blog<\/a> about one of our papers on neural networks for detecting malware, so I thought I'd link to it here. (You can read the paper itself, \"Malware Detection by Eating a Whole EXE\" here<\/a>.) Plus it was on the front page of Hacker News earlier this week, which is not something I thought would ever happen to my work.<\/p>\n

Rather than rehashing everything in Jon's Parallel for All post about our work, I want to highlight some of the lessons we learned from doing this about ML\/neural nets\/deep learning.<\/p>\n

As way of background, I'll lift a few paragraphs from Jon's introduction:<\/p>\n

The paper introduces an artificial neural network trained to differentiate between benign and malicious Windows executable files with only the raw byte sequence of the executable as input. This approach has several practical advantages:<\/p>\n