KAIST Professor Ki-Eung Kim of AI Graduate School Receives 'Influential Paper Award'

KAIST announced on the 10th that Professor Kim Ki-eung (photo) of the Kim Jae-cheol AI Graduate School received the Influential Paper Award from the International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS).

Professor Kim received the award for the paper titled "Cooperative Learning through Policy Search," which he co-authored and published in 2000.

Professor Kim Ki-eung. Provided by KAIST

In the paper, Professor Kim revealed that in cooperative learning among multiple AI agents in a distributed environment, the learning signals of individual agents do not depend on the information of other agents. Based on this, he proposed a distributed learning algorithm.

This learning algorithm is simple yet provides a guarantee of convergence to a local optimum. This explains why deep learning-based algorithms, which have become one of the main methodologies in multi-agent reinforcement learning research, have achieved remarkable success.

Professor Kim said, "I am deeply moved that the paper continues to be cited and utilized in multi-agent learning research using the latest deep learning techniques, and thanks to this, I received the 'Influential Paper Award' from IFAAMAS. I will continue to strive to produce impactful research results that can inspire future generations."

Meanwhile, IFAAMAS's "Influential Paper Award" was established in 2006 to select papers that have contributed to the field of autonomous agents and multi-agent systems research. IFAAMAS selects and awards 1 to 3 papers annually.