Opportunity | Call for Submissions: Cyber Autonomy Gym for Experimentation (CAGE) Challenge 3
Recent advances in artificial intelligence (AI) technologies show promise for autonomous cyber operations (ACO), offering the potential for distributed, adaptive defensive measures at machine speed and scale.
Cyber is a particularly challenging domain for autonomous AI. We nominate a challenge in this space that we believe requires further research in order to enable ACO to become an operational capability. To facilitate this AI research, we, the TTCP CAGE working group, are releasing our Cyber Operations Research Gym (CybORG), an experimental platform using the OpenAI Gym interface together with a cybersecurity scenario and a challenge to which we invite researchers to respond.
Our aim is to support the development of AI tactics, techniques and procedures with CybORG and a series of CAGE scenarios with associated challenge problems in order to support practical demonstrations of ACO. We wish to engage the AI and cyber security research communities, especially to leverage domain experts outside of the cyber field, and by encapsulating the cyber elements in environments such as CybORG along with the CAGE scenarios and challenge problems, we hope that the cyber problem set becomes accessible to a wider audience.
CAGE challenge 1 was released in August 2021 at the IJCAI-21 1st International Workshop on Adaptive Cyber Defense (ACD 2021), and concluded in in February 2022 with 4 novel submissions received. CAGE challenge 2 was released in April 2022, and concluded in July 2022 with 20 novel submissions received. CAGE Challenge 3 is now live! The CAGE scenario and associated challenge problem, together with CybORG (our AI gym research environment), are available here.
Any enquiries can be directed to PNTRNPBPunyyratr@qfg.qrsrapr.tbi.nh
Further instructions are included on the GitHub page.
30 September 2022: Challenge 3 released and open for submissions.
31 January 2023 (any time zone): Final day for challenge submissions.
14 February 2023: Final results announced.