This page was machine-translated and may differ from the original. View original
▲(Front row, left) Graduate student Taehyun Yoon and students from UNIST's Graduate School of Artificial Intelligence
New Lips Competition Student 1st Place, Global 2nd Place
UNIST AI Graduate School student Taehyun Yoon (advised by Professor Seongbin Lim) has won the top spot in the student category at a global artificial intelligence conference. This is a recognition of his achievement in raising the program’s performance to the highest level in a new field that has recently attracted much attention from the academic community.
UNIST announced on the 6th that Taehyun Yoon took first place among students and second place globally in the Primal Task category at the 'Machine Learning for Combinatorial Optimization (ML4CO)' hosted by 'NeurIPS', the most prestigious international conference in the field of artificial intelligence and machine learning.
The goal of this competition was to improve the performance of solver programs that solve combinatorial optimization problems with high computational complexity through machine learning. Combinatorial optimization is a field of industrial engineering and applied mathematics, and is used in industrial settings to find the most efficient methods for transportation routes and scheduling. Until now, algorithms designed by humans have been used primarily, but recently, attempts have been made to automate this based on data.
Student Taehyun Yoon said, “Earlier this year, I had the opportunity to look at related field papers in the lab, and I even implemented the unpublished code of the combinatorial optimization algorithm announced by Google DeepMind myself,” and “I heard that there was a competition, so I became interested and decided to participate.”
The competition was conducted in a way that periodically evaluated the performance of the program and ranked them. On the other hand, when the first results were released last July, Taehyun Yoon’s ranking was not high. He continued to try new things by repeatedly collecting data, implementing code, and testing them step by step.
Student Taehyun Yoon said, “I consistently worked hard with the thought of wanting to raise my ranking, and it started to become fun as my ranking gradually increased.” He continued, “I was ranked 4th globally before the final rankings were announced, and I was really happy when I saw that I had risen to 1st place among students and 2nd place globally in the final.”
This competition was held over four months from July to October, with 50 teams from companies and research institutes around the world competing to improve program performance. There were 23 double student teams.
Student Taehyun Yoon said, “I think the good results came from finding solutions to problems through fierce discussions with my research colleagues in the lab and not giving up and challenging myself,” and added, “I would like to give special thanks to my advisor, Dr. Hyo-Geun Yoon from Amazon, and Jin-Won Choi (Royce) from Kakao Enterprise for their advice and help during the competition.”
Professor Lim Seong-bin said, “It was an extremely competitive and challenging competition, and I am very happy and proud that the four months of hard work by the lab students have borne fruit.” He added, “I am proud that we were able to show the world the skills of UNIST students through this competition.”
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.

.png)













