About me
HI! I’m Doyeon Yoon, who likes math and is studying artificial intelligence. I mainly studied combinatorics in math and anti-bandwidth in graph structures.
I’ve been interested in a variety of fields, including finance, healthcare, and I can understand the key considerations in those fields.
Recently, I am interested in NLP, graph domain, and recommendation systems. I am studying with the next goal of achieving results through the combination of graph theory and deep learning. Thank you! 😍
Interests🙄
- Math : Graph theory, Anti-bandwidth
- DL/ML : Medical Segmentation, Financial
- Studying : NLP, GNN, Recommend System
Education🎓
- I received the M.S. degrees, in Department of Mathematics from Kwangwoon University, Seoul, Korea, in 2017.(Go to thesis)
- I received the B.S. degrees, in Department of Mathematics from Kwangwoon University, Seoul, Korea, in 2015.
ETC. Project 📚
AI·Big Data 인재양성 심화과정
포항공과대학교 인공지능연구원 PIAI(구 PIRL), Pohang
(2018.09 - 2018.10)
- Participated as a team leader on a machine learning project team, developing machine learning model, called a Senticle, which is a stock price prediction model using news data.
- NLP model using the ensemble of 1DCNN and LSTM(Go to code)
- Model description using LIME
부스트캠프 AI Tech
네이버 커넥트재단
(2021.01.12 - 2021.06.22)
- Participated in deep learning training, conducted 20 weeks of training and 4 contest-type project hands-on practice
- TIL repo
- [CV]Mask Image Classification(Go to code)
- [NLP]Relation Extraction(Go to code)
- [NLP]Dialogue State Tracking(Go to code)
- [KT]Deep Knowledge Tracing(Go to code)
Experience👨💻
Machine Learning Engineer
Hbee. Hwaseong-si, Gyeonggi-do
(2019.06.17 - 2020.03.13)
Worked on various machine learning related tasks, including but not limited to:
- Anomaly detection in semiconductor process.
- Stacked Auto-Encoder with short time series
- Semi-supervised GAN with image data
Research Intern
포항공과대학교 인공지능연구원 PIAI(구 PIRL), Pohang
(2018.11.01 - 2018.12.31)
Worked on various machine learning related tasks, including but not limited to:
- Medical image segmentation with 3D U-net(Go to code)
- Semi-supervised learning