Scholar
Taehwan Lee
Google Scholar ID: m-NupqQAAAAJ
UNIST
Citations & Impact
Citations
17
H-index
2
i10-index
1
Publications
5
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- - Publication: "Understanding Flatness in Generative Models: Its Role and Benefits," in International Conference on Computer Vision (ICCV), Honolulu, HI, 2025
- - Publication: "Benchmarking Federated Learning for Semantic Datasets: Federated Scene Graph Generation," Pattern Recognition Letters
- - Publication: "Rethinking the Flat Minima Searching in Federated Learning," in the 41st International Conference on Machine Learning (ICML), Vienna, Austria, 2024
- - Publication: "A new one-coincidence frequency-hopping sequence set of length p^2-p." 2018 IEEE Information Theory Workshop (ITW)
- - Award: "Rethinking the Flat Minima Searching in Federated Learning" was awarded as the finalist of Qualcomm Innovation Fellowship Korea (QIFK) 2024
Research Experience
- - Teaching Assistant: Electrical Engineering Programming Course EE233, Spring 2025 @ UNIST. Description: Basic programming tools for electrical engineering (C++)
- - Teaching Assistant: Electrical Engineering Programming Course EE233, Spring 2024 @ UNIST. Description: Basic programming tools for electrical engineering (C++)
- - ETRI Project (Development of a hierarchical and asynchronous federated learning simulation and contribution measurement tool, Jun. 2024 - Nov. 2024). Funded by: ETRI. My research focuses on coding the hierarchical and asynchronous structure of federated learning.
- - Korea-US (NSF) National Science Foundation International Joint Research (Dec. 2021 - Nov. 2024). Funded by: IITP. My part of the research is focused on developing an algorithm that is robust to the heterogeneity of clients in federated learning.
- - ETRI Project (Construction of a multi-inference performance-based federated learning asynchronous consensus experimental environment, Jun. 2023 - Nov. 2023). Funded by: ETRI.
- - ETRI Project (Development of technology to verify the feasibility of resource-hiding deep learning models, Apr. 2023 - Nov. 2023). Funded by: ETRI. My research focuses on adopting federated learning to Scene Graph Generation (SGG) tasks and constructing and analyzing the heterogeneity of the semantic dataset of Panoptic Scene Graph Generation (PSG) tasks.
Education
- - Ph.D., Artificial Intelligence Graduate School, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea. Advisor: Sung Whan Yoon (Lab: MIIT). Period: Mar. 2022 ~ Present
- - MS, Electrical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea. Advisor: Jinho Chung. Period: Mar. 2017 ~ Aug. 2019
- - B.S., Electronic Engineering, Hanbat National University, Daejeon, South Korea. Advisor: Kyoungjae Lee (Lab: WiSL). Period: Mar. 2011 ~ Feb. 2017
Background
- Research Interests: Federated and Decentralized Learning, Information and Communication Theory. Brief Bio: Has been a Ph.D. student at the Artificial Intelligence Graduate School (AIGS) at UNIST since March 2022.
Co-authors