Towards Automated Intelligent Operations with Machine Learning
The advent of Offensive Machine Learning (OML) is here, and while most offensive teams seem to be apprehensive about jumping on the hype train, we bought a first class ticket. When in a network, operators make decisions based on their aggregate experience and the information gathered from the network. To this end, operators are simply a filter for information - new information comes in and a command comes out.
The simplicity of the process is why automated offensive operations seems so approachable, yet current solutions are reduced to brittle logic that are not representative of a human decision making process. Machine Learning (ML) is the perfect candidate to model a non-linear decision making process. Can ML be used to assistant an operator? How about become one? In this talk we’ll tackle these challenges from the ground up and present a framework for building intelligent agents that model human operators.
ML + Ops = Love