본문 바로가기 메뉴 바로가기

Eight Research Articles from SNU College of Engineering Presented at Various International Conferences including SIGMOD and ICML

  • Uploaded by

    관리자

  • Upload Date

    2019.07.24

  • Views

    279

Eight Research Articles from SNU College of Engineering Presented at Various International Conferences including SIGMOD and ICML



▲ Professor Kunsoo Park, Professor Hyun Oh Song, Professor Gunhee Kim, Professor U Kang

 
SNU College of Engineering announced on 10th that Professor Kunsoo Park, Professor Hyun Oh Song, Professor Gunhee Kim, and Professor U Kang all from the Department of Computer Science and Engineering presented eight works to several international conferences on artificial intelligence and big data.
 
Professor Kunsoo Park published an article on partial graph matching algorithm to the 2019 ACM SIGMOD (Special Interest Group on Management of Data) Conference. The work introduced a new algorithm that searches for specific patterns in social networking services and it is known to be the fastest among all algorithms that perform the same function.
 
Professor Hyun Oh Song showcased his research findings on adversarial attacks on neural networks, reinforcement learning, unsupervised learning, and representation learning to two prestigious conferences. Two Long Talks and another article was sent to the International Conference on Machine Learning (ICML) and one was sent to the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
 
Professor Gunhee Kim also presented two research papers on auto-encoding and reinforcement learning to ICML. Through the two articles, Kim developed a framework named Variational Laplace Autoencoders for training deep generative models using the Laplace approximation and suggested the Curiosity-Bottleneck method that selectively analyzes new information from reinforcement learning based on the information bottleneck method.
 
Professor U Kang introduced his newest graph neural network technology, Belief Propagation Network (BPN), that sorts node in social networking services and webpages to the International Joint Conference on Artificial Intelligence (IJACI). The BPN that Kang suggested demonstrates great accuracy against conventional graph neural network.
 
SNU stated, “The international conferences that our professors from the Department of Computer Science and Engineering have submitted their research papers to all are leading conferences with authority and status in their fields. Thus, our professors presenting their works at these conferences demonstrate our leading research ability in artificial intelligence and big data.”