Research Interests

Machine learning and deep learning:

few-shot learning, meta-learning, multi-task learning, transfer learning

Education


KAIST, Daejeon, Korea

PhD., School of Computing

Sep 2022 - Present

M.S., School of Computing

Mar 2020 - Aug 2022

B.S., Department of Mathematical Science

Mar 2015 - Feb 2020

Publication


[1] Donggyun Kim, Jinwoo Kim, Seongwoong Cho, Chong Luo, Seunghoon Hong, "Universal Few-shot Learning of Dense Prediction Tasks with Visual Token Matching", In Proceedings of the International Conference on Learning Representations (ICLR), 2023. (notable-top-5%, oral presentation)

[2] Donggyun Kim, Seongwoong Cho, Wonkwang Lee, Seunghoon Hong, "Multi-Task Neural Processes", In Proceedings of the International Conference on Learning Representations (ICLR), 2022.

[3] Wonkwang Lee, Donggyun Kim, Seunghoon Hong, Honglak Lee, "High-Fidelity Synthesis with Disentangled Representation", In Proceedings of the European Conference on Computer Vision (ECCV), 2020.

Awards


Outstanding Paper Award in ICLR 2023

Mar 2023

29th Samsung Humantech Paper Award (silver prize)

Feb 2023

Qualcomm Innovation Fellowship South Korea (QIFK) 2022 Finalist

Oct 2022

Project Experience


Uncertainty Modeling and Calibration for Safe AI

KAIST, Korea

Mar 2020 - Mar 2021

Personalized Dialogue Systems

NCSoft Corporation, Korea

Mar 2019 - Dec 2019

Work Experience


SK hynix Inc., Incheon, Korea

Summer Internship, Data Science group

Jun 2018 - Aug 2018

Teaching Experience


Teaching Assistant

Undergraduate Tutor