About Me

Welcome to the homepage of Super Moon (nickname, Eng: Jackson.moon, Kor: 문종학)!

I'm a Ph.D student at Korea Advanced Institute of Science and Technology (KAIST) graduate school of AI in Korea.

Previous academic experience in medical sequence video data (Coronary Angiography) analysis through my Samsung Advanced Institute of Health Sciences & Technology (SAIHST) medical science master thesis and research experience in AI for healthcare through my Samsung Medical Center working years.

Now, my research interest is to develop practically useful AI models that can be predicted and analyzed like human.

In particular, multi-modal (Image and free-text) self-supervised learning mainly for radiology report generation. I am also interested in research using various medical data such as free-text, and Image.

The main theme of my ph.d course is a representation learning without supervision that can generally perform well on many disparate downstream tasks.

Applications (State-Of-The-Art Model):

Medical domain with visual-language multiomodality.

- Develop our model that can be generalized on 2 disparate downstream tasks. .(paper; Multi-modal understanding and generation for medical images and text via vision-language pre-training

Analysis (Theory-based unsupervised contrastive learning understanding):

General domain with visual modality.

- Understanding Dense Contrastive Learning. (paper; Correlation between Alignment-and-Uniformity and Performance of Dense Contrastive Representations)

Cite

@inproceedings{Moon_2022_BMVC,

author = {Jong Hak Moon and Wonjae Kim and Edward Choi},

title = {Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations},

booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},

publisher = {{BMVA} Press},

year = {2022},

url = {https://bmvc2022.mpi-inf.mpg.de/0844.pdf}

}

Contact Info:

Address: 8, Seongnam-daero 331beon-gil 18th floor KAIST AI, Bundang-gu, Seongnam-si, Gyeonggi-do, Republic of Korea

Email: jhak.moon@kaist.ac.kr, jhak.moon@gmail.com

Homepage for research projects detail: github.com/supersupermoon,

Homepage for CV: https://sites.google.com/view/super-moon

Linkedin: https://www.linkedin.com/in/super-moon