Network science is a field of study that aims to reveal structural patterns in real-world interaction networks, such as social networks, bibliographic networks, power grids, and others. The Network Science Lab (NS Lab; 네트워크과학연구실), part of the Dept. of AI at CUK, conducts extensive research on a wide range of theories, methodologies, and applications related to collection, representation, and analysis of networked data.
ProcGNN is a physics-informed graph neural network for predicting stream properties across chemical process flowsheets. It combines flow-aware bidirectional message passing with unit-level mass, component, and atom conservation regularization.
Van Thuy Hoang, who recently earned his doctoral degree from the Network Science Lab @ CUK, received the Excellence Award (우수상) at the 23rd Graduate School Academic Awards (제23회 대학원 학술상) Ceremony for his outstanding research in graph learning, following his Best Award (최우수상) at the 22nd awards last year.
The Network Science Lab at the Catholic University of Korea releases a Causality Inference Framework, namely CaMol, a novel causality inference framework for few-shot molecular property prediction.
The Network Science Lab at the Catholic University of Korea releases Multi-View Conditional Information Bottleneck, namely MVCIB, a novel pre-trained Graph Neural Network model specialised in 2D and 3D Molecular Structures.
We are pleased to announce the establishment of a new Major of AI-Chemical Engineering in the Department of Artificial Intelligence at the Graduate School, in addition to the existing Major of Artificial Intelligence.