Mingi Ji

 

Ph.D Candidate

Applied Artificial Intelligence Laboratory

Department of Industrial and Systems Engineering

KAIST, Daejeon, Republic of Korea

Contact: qwertgfdcvb [at] kaist.ac.kr / Mingi.Ji1015 [at] gmail.com

 

RESEARCH INTEREST

• Machine Learning
• Reinforcement Learning

 

EDUCATION

 

Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Sep. 2018 ~)

• Course of Doctor's Degree in AAILab, ISySE
• Academic Advisor: Prof. Il-Chul Moon

Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea (Sep. 2016 ~ Aug. 2018)

• Course of Master's Degree in AAILab, ISySE
• Academic Advisor: Prof. Il-Chul Moon

Seoul National University (SNU), Seoul, Republic of Korea (Mar. 2011 ~ Feb. 2016)

• Bachelor of Science in Naval Architecture and Ocean Engineering

PUBLICATIONS

 

International Conference

• Mingi Ji, Seungjae Shin, Seunghyun Hwang, Gibeom Park, Il-Chul Moon. Refine Myself by Teaching Myself : Feature Refinement via Self-Knowledge Distillation. Conference on Computer Vision and Pattern Recognition (CVPR 2021). Virtual Conference. June 19-25. [acceptance rate: 27%]
• Mingi Ji, Byeongho Heo, Sungrae Park. Show, Attend and Distill: Knowledge Distillation via Attention-based Feature Matching. AAAI Conference on Artificial Intelligence (AAAI 2021). Virtual Conference. Feb 2-9.
• Mingi Ji, Weonyoung Joo, Kyungwoo Song, Yoon-Yeong Kim, and Il-Chul Moon. Sequential Recommendation with Relation-aware Kernelized Self-Attention. AAAI Conference on Artificial Intelligence (AAAI 2020). New York. Feb. 7-12 [acceptance rate: 20.6%].
• Kyungwoo Song*, Mingi Ji*, Sungrae Park, and Il-Chul Moon. Hierarchical Context enabled Recurrent Neural Network for Recommendation. AAAI Conference on Artificial Intelligence (AAAI 2019). Hawaii. Jan. 27-Feb. 1 (* Equal Contribution) [acceptance rate: 16.2%].
• Sungrae Park, Kyungwoo Song, Mingi Ji, Wonsung Lee, and Il-Chul Moon, Adversarial Dropout for Recurrent Neural Networks. AAAI Conference on Artificial Intelligence (AAAI 2019). Hawaii. Jan. 27-Feb. 1  [acceptance rate: 16.2%].
• Seongcheol Woo, Juneyeong Yeon, Mingi Ji, Il-Chul Moon, Jinkyoo Park. Deep Reinforcement Learning with Fully Convolutional Neural Network to Solve An Earthwork Scheduling Problem. 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC). Miyazaki. Oct. 7-10.

Domestic Journal

• 지민기, 박준건, 김도형, 정요한, 박진규, 문일철. (2018). 강화학습을 이용한 이종 장비 토목 공정 계획. 한국시뮬레이션학회논문지, 27(1), 1-13.

Domestic Conference

• 지민기, 우성철, 박진규, 문일철. (2018). 강화학습을 이용한 다양한 환경에서의 절토 공정 계획. 한국경영과학회 학술대회논문집, 1805-1816.
• 지민기, 우성철, 정요한, 박진규, 문일철. (2017). 딥러닝 기반 강화학습을 이용한 토목 공정 계획. 대한산업공학회 추계학술대회 논문집, 1485-1511.

 

LAB 

   Mingi Ji is a member of Applied Artificial Intelligence Laboratory. (2017 ~)