Multi-agent Reinforcement Learning

A single-agent reinforcement learning has achieved great success. However, WE NEVER LIVE ALONE! Our next mission is building reinforcement learning algorithms for multi-agent systems with cooperative and/or competitive strategies. This project is an essential step to build the society of artificial intelligence, which includes many different kinds of engineering problems (e.g, mobile robot networks, autonomous driving technologies). Our solutions are motivated by game theory, cognitive neuroscience, and network science.

AI for Future Networks

Recent advances in artificial intelligence and machine learning have been embedded in all forms of technology. Now, it is time to change networks! AI can offer end-to-end learning to lead to higher throughput, lower latency, and power consumption. A data-driven model can design an intrusion detector that is able to catch even unseen types of malicious traffic.

Brain Science (Computational Neuroscience)

Life is a sequence of decision-making events. We, as a species of animals, make decisions every day, every moment. Then, how do “we” make decisions? Do we always find the “optimal” solution? These are fundamental scientific questions to understand how the brain works and how to build an algorithm for artificial intelligence.

Network Science

Our environment including social, cognitive, biological and engineering systems can be modeled as a complex network. For example, the spread of Coronavirus disease (COVID-19) can be explained and analyzed using network science. In this project, we aim to analyze the complex network, extract an important meaning, and help the world.

Active Research Grants

  • Development of R&D Human Resource and Technologies for Intelligent Cyber Threat Responses
    • Institute of Information & Communications Technology Planning & Evaluation (IITP)
    • 2020.07.01-2025.12.31
  • Development of Distributed/Cooperative AI based 5G+ Network Data Analytics Functions and Control Technology
    • Institute of Information & Communications Technology Planning & Evaluation (IITP)
    • 2021.04.01-2025.12.31
  • Stability and Divergence Analysis for Offline Reinforcement Learning Based Autonomous Vehicles in Mixed-autonomous Traffic
    • Hyundai Motor Company (Hyundai NGV)
    • 2022.12.26-2023.08.25
  • Development of Machine Learning Algorithms for Personal Identification in Multiplexed Single Cell Transcriptomes
    • National Research Foundation of Korea
    • 2023.03.01-2024.02.28

Completed Research Grants

  • Reinforcement Learning Based Distributed Control for Mobile Networks
    • National Research Foundation of Korea
    • 2020.06.01-2023.02.28
  • Supervised Agile Machine Learning Techniques for Network Automation based on Network Data Analytics Function
    • Institute of Information & Communications Technology Planning & Evaluation (IITP)
    • 2020.03.01-2021.12.31
  • Distributed Network Formation Strategy Based on Reinforcement Learning
    • Soongsil University
    • 2020.03.01-2021.02.28

%d bloggers like this: