Anonymous 09/08/2017 (Fri) 14:21:11 No. 11052 del
>>11029
>In this work we explore the feasibility of applying neural networks to malware detection and feature learning.
buzz buzz buzzword. why do we care about some minor innovations to goy tech? the single one justification for antivirus to exist is that it purges outbreaks on machines owned by the goy who run outdated software and open untrusted executables (these people will have already have had their info stolen by the time the antivirus is updated to have a signature for such malware). it does not offer any security or any useful attribute to anyone who knows what they're doing. in particular, using NN to "detect malware" is just another heuristic full of false positives and will be once again, trivial to bypass.
>>11035
this
>>11037
PDF is worse than HTML in every way
inb4 >muh typography!11
>>11048
>Fortunately (or unfortunately if you refuse to learn) ML you'll be left behind as a metasploit monkey and etc
is this what people conclude when they reason on the level of memes?
>AVs check for warez
what did he mean by this
>It seems like a much more efficient process than sandboxing a suspected snippet and running a sim.
confirmed for not having a clue what you're talking about