PSYCHIC: Predictive System for Cyber Hostility using Integrated Computational Models
Cyberattacks are a growing problem that cause over $300 billion in system recovery and data losses each year. Predicting attacks before they occur would be a decisive advantage. The Intelligence Advanced Research Projects Activity (IARPA) provided the opportunity to create automated methods to forecast and detect attacks under the Cyberattack Automated Unconventional Sensor Environment (CAUSE) program. As part of this program, Charles River led a team to search for leading indicators of possible impending attacks.
“Under the PSYCHIC effort, we’re predicting cyberattacks against particular targets. The goal is to predict when a company or organization might be attacked and how, which enables them to take defensive measures ahead of time.”
Senior Scientist and Principal Investigator for the PSYCHIC effort
Charles River is led an interdisciplinary team of computer scientists, cyber security experts, cognitive scientists, natural language processing experts, and human factors experts across industry and academia. The team included BBN Technologies, Battelle, Assured Information Security, the University of Utah, Arizona State University, and Professor Alexander Levis of George Mason University.
Charles River will drew on its expertise in ensemble modeling, natural language processing, cyberattack modeling, and predictive analytics in the PSYCHIC effort. In part, the PSYCHIC solution built upon malware models developed in the DARPA-funded MAAGI effort to predict the possible evolution of attacks and where they might occur. Charles River also used its cyber modeling and reactive agent framework, CyMod, to predict idiosyncratic patterns of different kinds of cyberattacks.
“We are excited to work with this talented team on this challenging and important problem,” said Dr. Scott Neal Reilly . “And we are happy to work on a new program with IARPA.”
Dr. Scott Neal Reilly, Senior Vice President
PSYCHIC is Charles River’s most recent program effort with IARPA. Earlier this year, Charles River was awarded a contract as part of IARPA’s SCITE program to help predict the performance of enterprise-level analytic systems, and 2014, it was awarded a contract to lead a team to improve adaptive reasoning and problem solving under the SHARP program.
Predictive System for Cyber Hostility using Integrated Computational Models (PSyCHIC) is supported by the Office of the Director of National Intelligence (ODNI) and the Intelligence Advanced Research Projects Activity (IARPA) via the Air Force Research Laboratory (AFRL) contract number FA8750-16-C-0108. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon.
The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of ODNI, IARPA, AFRL, or the U.S. Government.