Perspecta Labs—the innovative applied research arm of Perspecta Inc.—has been awarded a defensive cyber-contract potentially worth $14.5m by the United States Army.
The lucrative contract was handed out by the US Army Combat Capabilities Development Command (CCDC), Command, Control, Computers, Communications, Cyber, Intelligence, Surveillance and Reconnaissance Center (C5ISR), Space and Terrestrial Communications Directorate.
Perspecta Labs has been chosen by C5ISR to work on the Autonomous Defensive Cyber Operations program. The objective of the program is to create a suite of autonomous cyber-capabilities for the US Army that are adaptable, secure, and resilient to any deceptive tactics practiced by America's adversaries.
Another key feature of the cyber-solution that Perspecta Labs has been tasked with developing is that it must be easy to configure and deploy.
Perspecta Lab's role within the program will be to research, design, develop, demonstrate, and deliver a machine learning for defensive cyber-operations solution. This tool will be used for training, rapidly deploying, and retraining containerized cyber-sensors that detect both new and previously discovered vulnerabilities, attacks, and malware.
The aim is to come up with an automated solution that can respond to threats much faster than a system that relies on human intervention.
"Perspecta Labs will leverage its extensive expertise in machine learning, cybersecurity and tactical networking to provide innovative and effective autonomous defensive cyber operations to the Army," said Perspecta Labs president Petros Mouchtaris.
"We are honored to have been selected to develop a solution that meets the Army's tactical needs for autonomy, security, ease of use, adaptability, efficiency and robustness."
Perspecta Labs said that the solution it develops will utilize the company's novel machine learning paradigm, LUPI (Learning Using Privileged Information) to "create highly accurate cyber sensors in a heterogeneous environment."
These cyber-sensors are not only resilient to adversarial machine learning attacks but also can be dynamically retrained to take into account the large variations present in the tactical environment.
A spokesperson for Perspecta Labs said: "The solution's architecture harnesses these cyber sensors to detect diverse attacks and incorporates a cognitive agent for automatically generating courses of action for the cyber-defender."