Byun, Hyeongmin (변형민)

[first-name-in-lower-case].[last-name-in-lower-case]@gmail.com

Working Experience

Moloco, Seoul, Korea

• Applied Scientist / Oct 2024 - Present

Hyperconnect, Seoul, Korea

• Data Scientist / Mar 2022 - Oct 2024

• Machine Learning Software Engineer / Nov 2019 - Mar 2022

• Machine Learning Engineer / Mar 2018 - Nov 2019

Education

KAIST Business School, Seoul, Korea

• Master of Engineering, Managerial Economics, School of Management Engineering / Feb 2016 - Feb 2018
  • Research Theme: Business Cycles, Macro-Finance and Macroeconometrics
  • Thesis: Effects of Investment-Specific Technology Shocks in Korea. Accessible via KAIST Library

Seoul National University, Seoul, Korea

• Master student, Statistical Learning & Computational Finance Lab, Department of Industrial Engineering / Mar 2014 - Jun 2015 (Suspended)
  • Research Theme: Analysis of Financial Derivatives with Machine Learning Models
• Bachelor of Science in Engineering, Department of Industrial Engineering / Mar 2009 - Feb 2014
  • with double major in Mathematical Science; double major in Statistics; and minor in Economics. Graduated Cum Laude

Seoul Science High School, Seoul, Korea

• Mar 2006 - Feb 2009

Honors and Awards

LPIRC (Low Power Image Recognition Challenge), CVPR 2018 Conference

• Second Prize for Track 1 with Hyperconnect ML Team, June 2018.

Dean’s List, KAIST Business School

• High Honor for Master in Management Engineering, Spring 2016

University SCM Competition, Korea Society of Supply Chain Management

• First Prize with Soovin Yoon, Seok-joo Kwon and Prof. Il-kyung Moon, June 2013.
  • Topic: Developing Smartphone Recycling Model using SCM
  • news article

Lee Jung-Han Award, Department of Industrial Engineering, Seoul National University

• Runner-Up for Academic Division, Undergraduate, 2012. link

University Mathematics Competition, Korea Mathematics Society

• Gold Prize (ranked 2-14th) in Mathematics major division / Nov 2012
• Gold Prize (ranked 2-10th) in Mathematics major division / Nov 2011
• First Place in non-Math major division / Oct 2010

Publications and Conference Presentations

• Seong Min Kye, Kwanghee Choi, Hyeongmin Byun and Buru Chang*. “TiDAL: Learning Training Dynamics for Active Learning”. Arxiv 2210.06788 link; International Conference on Computer Vision 2023, Oct 2023, Paris, France link
• Seungwoo Choi, Seokjun Seo, Beomjun Shin, Hyeongmin Byun, Martin Kersner, Beomsu Kim, Dongyoung Kim* and Sungjoo Ha*. “Temporal Convolution for Real-time Keyword Spotting on Mobile Devices”. Arxiv 1904.03814 link; Interspeech 2019, Oct 2019, Graz, Austria.
• Seokjun Seo, Seungwoo Choi, Martin Kersner, Beomjun Shin, Hyungsuk Yoon, Hyeongmin Byun and Sungjoo Ha*. “Towards Real-Time Automatic Portrait Matting on Mobile Devices”. Arxiv preprint 1904.03816 link
• Youngdoo Son, Hyeongmin Byun and Jaewook Lee*. Nonparametric Machine Learning Models for Predicting the Credit Default Swaps: An Empirical Study. Expert Systems with Applications, Volume 58, 1 October 2016, Pages 210-220, ISSN 0957-4174. link
• Hyeongmin Byun, Youngdoo Son and Jaewook Lee. Effects of Keyword Search Volume on Credit Default Swap Spread Prediction. KIIE (Korea Institute of Industrial Engineers) Conference, April 2015. Jeju, Korea. link
• Youngdoo Son, Hyeongmin Byun and Jaewook Lee. “Nonparametric Models for Predicting Credit Default Swap Prices: Empirical Study.” INFORMS (Institute for Operations Research and Management Science) Annual Meeting 2014, Nov 2014. San Francisco, CA, USA.
• Youngdoo Son, Saerom Park, Hyeongmin Byun and Jaewook Lee. Computing Default Probability using Ensemble Method. 15th APIEMS (Asia Pacific Industrial Engineering and Management Systems) Conference, Oct 2014. Jeju, Korea. Best Student’s Paper Award news

Writings and Blog Articles

• 아자르에서 AI 기반 추천 모델의 타겟 지표를 설정하는 방법 (feat. 아하 모멘트) with Joonyoung Yi and Hyeokhwa Seong, Hyperconnect Tech Blob, Apr 2024
• Tips for building fast portrait segmentation network with TensorFlow Lite with Hyperconnect ML Team, Hyperconnect Tech Blog, Jul 2018
• Low Power Image Recognition Challenge 2018 with Hyperconnect ML Team, Hyperconnect Tech Blog, Jun 2018