HAICon2021 산업제어시스템 보안위협 탐지 AI 경진대회

알고리즘 | 정형 | 이상탐지 | 제어 | TaPR

  • moneyIcon Prize : 총 2,000만원
  • 878명 마감

[Title]

  • AI Challenge for Industrial Control System Threat Detection


[Subject]

  • Industrial Control System Security Threat Detection


[Background]

  • Recently, cybersecurity threats to the control systems of national infrastructure and industrial facilities are continuously increasing. Cyberattacks on critical national facilities can cause enormous irreparable damage to countries and societies, and they are concentrating on developing security technologies to prevent the threats.
  • The industrial control system security dataset that has the characteristics of on-site control systems and includes various control system cyber attacks is an essential element for AI-based security technology research.
  • In response to the needs of these research sites, the National Security Research Institute established a testbed for a power generation control system using industrial control devices, sensors, and actuators from GE, Emerson, and Siemens in 2019.
  • The National Security Research Institute is continuously developing industrial control system security data sets (HAI) using the testbed and has released HAI 20.07 and HAI 21.03 data sets to solve data shortages in research sites.
  • HAICon 2021 will be held for the second time following the first competition HAICon 2020 to activate AI-based security threat detection research and dissemination of technology using HAI security dataset.
  • We intend to continuously raise the level of related skills by sharing the model of the winning team, and an improved version of the competition dataset will be released through feedback from participants.


[Host/Organizer/Support/Operation]

  • Host and organizers: National Intelligence Service and National Security Research Institute, South Korea
  • Support: Korea Institute of Information Security & Cryptology, South Korea
  • Operation: Dacon


[Eligibility for Participation]

  • Any Korean can participate.
  • Foreigners must team up with Koreans.
  • Organizations (universities, education companies, etc.) should follow the group application process


[Reference Guide]

  • You should cite the following papers in your paper with the HAI security dataset or the eTaPR evaluation tool.

- HAI 21.03, HAICon 2020

Hyeok-ki Shin, Woomyo Lee, Jeong-Han Yun, and Byung-Gil Min, "Two ICS Security Datasets and Anomaly Detection Contest on the HIL-based Augmented ICS Testbed", Workshop on Cyber Security Experimentation and Test, 2021.

- HAI 20.07

Hyeok-ki Shin, Woomyo Lee, Jeong-Han Yun, and Hyoungchun Kim, "HAI 1.0: HIL-based Augmented ICS Security Dataset", Workshop on Cyber Security Experimentation and Test, 2020.

- TaPR

Won-seok Hwang, Jeong-Han Yun, Jonguk Kim, and Hyoungchun Kim, "Time-Series Aware Precision and Recall for Anomaly Detection - Considering Variety of Detection Result and Addressing Ambiguous Labeling", CIKM'19: Proceedings of the 28th ACM International Conference on Information and Knowledge Management, 2019.


※ HAICon's evaluation tool, eTaPR, is an upgraded version of TaPR.

Main Event Schedule

  1. 08.02

    Start of team registration &
    warm-up

  2. 08.13

    End of warm-up

  3. 08.17

    Start of HAICon2021

  4. 10.15

    Team Merger Deadline

  5. 10.22

    Test submission deadline

  1. 10.27

    Code submission deadline

  2. 11.12

    Final result announcement

[Title]

  • AI Challenge for Industrial Control System Threat Detection


[Subject]

  • Industrial Control System Security Threat Detection


[Background]

  • Recently, cybersecurity threats to the control systems of national infrastructure and industrial facilities are continuously increasing. Cyberattacks on critical national facilities can cause enormous irreparable damage to countries and societies, and they are concentrating on developing security technologies to prevent the threats.
  • The industrial control system security dataset that has the characteristics of on-site control systems and includes various control system cyber attacks is an essential element for AI-based security technology research.
  • In response to the needs of these research sites, the National Security Research Institute established a testbed for a power generation control system using industrial control devices, sensors, and actuators from GE, Emerson, and Siemens in 2019.
  • The National Security Research Institute is continuously developing industrial control system security data sets (HAI) using the testbed and has released HAI 20.07 and HAI 21.03 data sets to solve data shortages in research sites.
  • HAICon 2021 will be held for the second time following the first competition HAICon 2020 to activate AI-based security threat detection research and dissemination of technology using HAI security dataset.
  • We intend to continuously raise the level of related skills by sharing the model of the winning team, and an improved version of the competition dataset will be released through feedback from participants.


[Host/Organizer/Support/Operation]

  • Host and organizers: National Intelligence Service and National Security Research Institute, South Korea
  • Support: Korea Institute of Information Security & Cryptology, South Korea
  • Operation: Dacon


[Eligibility for Participation]

  • Any Korean can participate.
  • Foreigners must team up with Koreans.
  • Organizations (universities, education companies, etc.) should follow the group application process


[Reference Guide]

  • You should cite the following papers in your paper with the HAI security dataset or the eTaPR evaluation tool.

- HAI 21.03, HAICon 2020

Hyeok-ki Shin, Woomyo Lee, Jeong-Han Yun, and Byung-Gil Min, "Two ICS Security Datasets and Anomaly Detection Contest on the HIL-based Augmented ICS Testbed", Workshop on Cyber Security Experimentation and Test, 2021.

- HAI 20.07

Hyeok-ki Shin, Woomyo Lee, Jeong-Han Yun, and Hyoungchun Kim, "HAI 1.0: HIL-based Augmented ICS Security Dataset", Workshop on Cyber Security Experimentation and Test, 2020.

- TaPR

Won-seok Hwang, Jeong-Han Yun, Jonguk Kim, and Hyoungchun Kim, "Time-Series Aware Precision and Recall for Anomaly Detection - Considering Variety of Detection Result and Addressing Ambiguous Labeling", CIKM'19: Proceedings of the 28th ACM International Conference on Information and Knowledge Management, 2019.


※ HAICon's evaluation tool, eTaPR, is an upgraded version of TaPR.

Main Event Schedule

  1. 08.02

    Start of team registration & warm-up
  2. 08.13

    End of warm-up
  3. 08.17

    Start of HAICon2021
  4. 10.15

    Team Merger Deadline
  5. 10.22

    Test submission deadline
  6. 10.27

    Code submission deadline
  7. 11.12

    Final result announcement