Department of Statistics and Data Science, University of Seoul
Optim.Lab. at the Department of Statistics and Data Science, University of Seoul, studies the fundamental principles of machine learning and develops optimization methodologies for learning models.
Our research aims to understand how learning algorithms operate, analyze their theoretical properties, and design effective optimization strategies for model training and generalization.
We primarily work with:
We build knowledge graphs from complex relational data in domains including molecules, finance, and documents, and use graph neural networks and language models for prediction, inference, fact checking, and discovery.
- Methods:
Heterogeneous Graph Embedding,LLM,Agentic AI
We study information retrieval and recommendation with knowledge graphs, large language models, and multimodal data, including bias correction, reliability, and personalization.
- Methods:
Causal Inference,Recommendation Algorithm,Query Performance Prediction,Task-Oriented Dialogue System,Fact Verification
We study generative models for heterogeneous tabular data, including relationships among tables, schemas, and temporal changes.
- Methods:
Variational AutoEncoder,Generative Models
We study the prediction of spatial and temporal changes in Arctic sea ice and the likelihood of ice opening, based on its physical and thermodynamic properties.
- Methods:
Spatiotemporal AI
You can explore our research publications via Google Scholar:
