Changhun Kim
Changhun Kim, M. Sc.
Research Project: GridAssist
Academic CV
Education
- M.S. in Artificial Intelligence, Friedrich-Alexander University Erlangen-Nürnberg, Germany (Sep 2022 – Mar 2025)
- Thesis: RetNetHTR – Leveraging Retentive Networks for Efficient and Accurate Handwritten Text Recognition
- Teaching Assistant: Advanced Deep Learning, Introduction to Machine Learning
- Research Assistant, Medical Imaging AI Lab
- B.S. in Computer Science, University of Seoul, South Korea (Mar 2014 – Feb 2021)
- Undergraduate Research Assistant, Environmental System Toxicology Lab
- Two-year absence due to mandatory military service
Experience
- Research Assistant, AI Medical Imaging Lab, FAU Erlangen-Nürnberg
- Project: Self-supervised denoising of fluorescence microscopy images
- Researcher, Urban Big Data and AI Institute, University of Seoul (Jan 2021 – Sep 2022)
- Project: Graph Neural Networks for molecular toxicity prediction
- AI Educational Content Technical Quality Assurance , MODULABS, South Korea (May 2021 – Oct 2021)
- Research Intern, Environmental System Toxicology Lab, University of Seoul (Sep 2018 – Dec 2019)
- Project: Machine learning for imbalanced toxicity data prediction
Running Thesis
Type | Title | Status |
---|---|---|
Project | A Resource-Efficient AC Power Flow Prediction Framework using Physics-Informed GNNs and RL-Based Model Compression | running |
MA thesis | [Thesis] Reinforcement Learning for 110 kV Distribution Grid Restoration in Blackout situation | running |
Projects
No projects found.
Research Project: GridAssist – AI-based techniques for automated congestion control and fault mangement in power grids, while also buidling a robust knowledge database that integrates and harmonizes heterogeneous power grid data from multiple sources.
Research Areas: Congestion and Fault mangement in Electrical Grid, Graph Neural Networks, Time Series Data Analysis, and Reinforcement Learning for Decision-Making in Power Systems.
Publications
No publications found.
Lectures
No matching records found.