Prototype Orchestration Framework as a High Exposure Dimension Cyber Defense Accelerant Amidst Ever-Increasing Cycles of Adaptation by Attackers
share
The cycles of adaptation by attackers are ever-increasing. To meet these evolving threats, outsourcing to Managed Security Service Providers (MSSPs) has become prevalent. As these MSSPs contend with a torrent of varied attack vectors, they are increasingly utilizing Artificial Intelligence (AI) to assist them in protecting their clients. Practitioners often assert that systems which provide decisions can be construed as AI; along this vein, this paper presents summary results of a prototype orchestration framework that selects and prioritizes cyber tools to be utilized against a continuous stream of testbed cyber-attacks. This orchestration framework is predicated upon the hybridization of a modified Deep Belief Network (DBN) conjoined with a particular cognitive computing precept (the acceptance of higher uncertainty amidst lower ambiguity for compressed decision cycles); for uncompressed decision cycles, it utilizes a modified Stacked Generative Adversarial Network (SGAN), which serves as a feeder to a Lowering Ambiguity Accelerant (LAA). Results show promise during the 1-5 day period; work has already commenced for improving the performance for day 6+, and uptime is already at 38 days with minimal degradation.
READ FULL TEXT