20 May 2016

AI + Humans = AI2 for Cyber security

Cyber security is the major challenge of today's and tomorrow’s world.

A group of researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) are working with machine-learning startup PatternEx to develop an Artificial Intelligence system that can detect 85 percent of attacks by reviewing data from more than 3.6 billion lines of log files each day and informs anything suspicious.

In this interesting paper the authors present their analyst-in-the-loop security system, which combines analyst intuition with state-of-the-art machine learning to build an end-to-end active learning system. They have called this system AI2 (AI Squared) as it  learns and automatically creates models that, when executed on new data, produce predictions as intelligent as those deduced by human analysts. In other words the human analyst then identifies which events are actual cyber attacks and which aren't. This feedback is then incorporated into the machine learning system of AI2 and is used the next day for analyzing new logs. The system has four key features: a big data behavioral analytics platform, an ensemble of outlier detection methods, a mechanism to obtain feedback from security analysts, and a supervised learning module.

A video here: http://thehackernews.com/2016/04/artificial-intelligence-cyber-security.html

Structural similarities between future Internet and the "nervous system" of a living entity indeed may help to develop enhanced AI systems in the Web, combining humans capabilities, pervasive sensors and actuators with state-of-the-art machine learning to build an end-to-end active learning and pervasive system.

Optimizing any socio-economic processes (precision agriculture, I4.0, business agility), searching, sorting, and interpreting data and information, and above all cyber security will be the “killer application” for this WWW AI2: maybe billions of people would be willing to pay for it!