Report Available! Experiment and improve reinforcement learning algorithms to enhance anomalous network behaviour detection

Cybersecurity is a significant research area because all of the operations based on government, military, commercial, financial and civilians gather, process, transfer and store tremendous volume of data on computers and others. Cyber-attacks have imposed increasing threats and damages on our modern society at all levels. Network Intrusion Detection System (NIDS) is one of the major techniques in preventing cyber-attacks occurred in network traffic. Over the past decade, a lot of research work has been conducted to explore the capabilities of artificial intelligence (AI) methods in developing NIDS solutions. The previous studies suggested that AI algorithms have promising potentials in developing effective solutions to detect the increasing attacks.

TeleMARS R&D team commits to advance AI-based methods, explore realistic approaches of deploying the research outcomes in real network environment, and support on-going research in wider community to achieve long term sustainable development. The key objectives of this project were to:

  • contribute to the development of NIDS;
  • contribute to research community in the subject of anomaly detection;
  • establish a practical collaboration framework to enable scientists and IT professionals from diverse background to work together to continuously contribute to NIDS research;
  • test and prove TeleMARS operation and technical frameworks, and the team capabilities; and
  • inspire and enable the participation of broader research community in cybersecurity domain supporting gender equality and inclusion

This project started in September 2020 and finalized in June 2021. The main activities included:

  • Literature review and project design.
  • Data analysis and preparation.
  • Anomaly detection model development using Machine Learning methods including Reinforcement Learning method.
  • Model experimentation.
  • Established evaluation pipelines to simulate real application environment.
  • Model capability evaluation applying different datasets.
  • Implementation of a collaboration framework supporting the research activities conducted by researchers and professionals with various backgrounds.

The final technical report is available for review here.

You Might Also Like

Published by

Sylvia Cadena - ISIF Asia secretariat

After ten years managing the Information Society Innovation Fund (ISIF) at APNIC, Sylvia was appointed Head of Programs at the APNIC Foundation in December 2016. As Head of Programs, Sylvia works on the management, design and implementation of collaborative programs to expand APNIC’s capacity building efforts and its overall Development Program. At ISIF Asia, Sylvia continues to lead a grants and awards program that seeks to empower communities in the Asia Pacific to research, design and implement Internet-based solutions for their own needs. One of the region’s most established Internet development programs, ISIF places particular emphasis on the positive role the Internet has on social and economic development in the Asia Pacific. Sylvia also leads APNIC’s engagement with the Seed Alliance, a global collaboration with the African (AFRINIC) and South American (LACNIC) Internet registries. Together with its funding partners, and various regional sponsors, the Seed Alliance supports innovative Internet development across the global south. She is also a member of the ICANN CCWG on new gTLD auction proceeds and co-chair of the APrIGF. Throughout her career, Sylvia has focused on the strategic use of the Internet for development with an emphasis on capacity building. Since her early years as a UN Volunteer, she has worked across the multi-stakeholder spectrum of organizations with technical and advisory roles, mainly about information systems, access provision and innovation. In July 2003, her work was recognized with the "Annual Award for Young Professionals" by the International Development Research Center (IDRC).