Deep Learning App Targets Malware

Be perfect,be yourself.

Here's the problem with most programs aimed at killing malicious software: 

They need someone to tell them something's malicious.

What if, however, the programs had the smarts to identify bad code on their own? That's what a company called Deep Instinct says its security solution, launched last week, can do.

The offering works its magic with a technology called "deep learning."

"Deep learning draws its inspiration from the human mind. It organizes itself into a structure of synthetic neurons," explained Bruce Daley, principal analyst at Tractica.

"It's another term for neural networks," he told TechNewsWorld. "It was rebranded because there was so little progress with neural nets.

Better Brain Emulation

Deep learning applications will be a hot commodity in the future, becoming a $10 billion market by 2024, Tractica forecasts.

A fertile area for those apps may be security, which is what Deep Instinct is counting on.

Classical neural networks in the 1980s and '90s had one or two layers of several hundred neurons.

"Nice results, but nothing spectacular," observed Deep Instinct CTO Eli David.

Now with advances in hardware, processing power and algorithms, deep neural networks that are more than 10 layers deep with hundreds of millions of neurons can be created. That kind of power can be harnessed to approach software development in a different way