MIT Lincoln Laboratory helps inaugurate the Workshop on Artificial Intelligence and Cyber Security

by Nathan Parde

The first-ever workshop on Artificial Intelligence for Cyber Security (AICS) was held on 12 February as part of the Association for the Advancement of Artificial Intelligence's (AAAI) annual conference in Phoenix, Arizona. Members of MIT Lincoln Laboratory's Cyber Security and Information Sciences Division—David Martinez, William Streilein, and Kevin Carter—teamed with Arunesh Sinha, a colleague from the University of Southern California, to organize and chair the workshop.

"AICS assists analysts in dealing with the deluge of cyber data to more rapidly reach decisions and actions, enabling us to keep pace with the rapid increase in cyber threats," said Martinez, associate head of the Cyber Security and Information Sciences Division.

A panel discussion wrapped up the 2016 Association for the Advancement of Artificial Intelligence's Workshop on Artificial Intelligence for Cyber Security.

James R. Clapper, U. S. Director of National Intelligence, has called the intelligent cyber adversary the number one threat to U.S. security. Accordingly, Lincoln Laboratory has named cyber security and autonomous systems as two of the top thrusts for its own research agenda. "The AICS workshop aligns directly with these Laboratory mission areas and aims to bring together researchers and practitioners from both fields in an environment that fosters collaboration and furtherance of novel capability development," said Streilein, who leads the Cyber Analytics and Decision Systems Group at the Laboratory.

The workshop focused on ways to improve cyber security capabilities through the research and application of artificial intelligence technologies, including machine learning, game theory, natural language processing, knowledge representation, and automated and assistive reasoning. "The workshop is unique in that it specifically approached the cyber security problems from an AI perspective," explained Streilein. "Solutions were considered that would incorporate machine assistance and intelligence from the outset, and therefore have the best promise of dealing with problems of volume, velocity, variety, and veracity—the big data problems that go beyond human capabilities."

The workshop featured two invited keynote speakers, a panel discussion, and nine peer-reviewed papers on topics related to the cyber threat, human-machine interaction, and cyber situational awareness.

George Cybenko, a professor in the Thayer School of Engineering at Dartmouth College, opened the workshop with a keynote address on the application of AI to cyber security problems. In his afternoon keynote presentation, Robert Laddaga, a professor of electrical engineering and computer science at Vanderbilt University, reviewed the future of AI and cyber security. "Dr. Laddaga’s keynote was interesting for its vision of the future of cyber security, which will inevitably include AI solutions in response to an adversary who is leveraging AI for nefarious purposes," said Streilein.

The workshop concluded with a panel discussion that focused on the impediments to acceptance of AI within the cyber security community. Among the panel experts were Richard Lippmann, an MIT Lincoln Laboratory Fellow; Vern Paxson, a professor at the University of California, Berkeley; Milind Tambe, a professor at the University of Southern California; and Benjamin Rubinstein, a researcher from the University of Melbourne. "Each panelist brought a different perspective to the problem, yet they found many common challenges and recommendations for paths forward," said Streilein, who served as moderator for the panel. While the panel determined there is a place for AI in cyber security operations, they cautioned against using AI without fully understanding it.

Because of the 2016 workshop's success, the AAAI conference committee asked that the AICS workshop be repeated next year, perhaps with an expanded agenda. Potential themes for next year include a special focus on human-in-the-loop systems and lessons learned from analogous research areas, such as biology.

Looking forward, the AICS workshop will continue to bring together diverse areas of expertise to help deal with the rapidly evolving cyber security landscape, said Streilein, who added, "As workshop chairs, we hope visitors left with a sense that cyber security is an important national concern that needs novel solutions to augment human capabilities and that AI can have an important role to play in creating solutions."

Posted March 2016

top of page