- May 16, 2018
- 1,363
I found some good content on blog of NortonLifeLock research team
![]()
Machine Learning for Security: Things Will Work Better with Humans in the Loop
Why Machine Learning for Security Is Difficultwww.nortonlifelock.com
From the article:
Bottom line: When it comes to cyber security, ML is going to be more effective when it is applied in conjunction with human experts in the loop, rather than working on its own. That’s where you can talk about a complete solution. But it needs humans to be fully effective.
Yeah, this makes perfect sense -- and kinda what I've been saying for quite a while.
I ran Cylance for a while with Norton (and other things).
I've always thought that Cylance should be the ML Module of a suite rather than a stand-alone product.
At least at one time, the ML algorithm of Cylance was the real deal. But Cylance just had too many false positives. It was a pain-in-the-ass. @Nightwalker lost his patience with Cylance before I did.... but eventually, I lost my patience too.
Cylance is still on multiple machines (I seem to have gotten a free long term license based on where I work). But I keep it turned off mostly.