Jaehyung Kim

I am a 4th year Ph.D. student at ALIN Lab, advised by Prof. Jinwoo Shin at Korea Advanced Institute of Science and Technology (KAIST). I received an M.S. in Electrical Engineering (Feb. 2019) and a B.S. in Mathematics (Feb. 2017) in KAIST.

My research goal is to build an effective machine learning (ML) pipeline to tackle various real-world problems. To this end, I have been focused on developing improved ML algorithms for applying it to more practical and challenging scenarios, e.g., low-resource, class-imbalanced, and semi-supervised learning. Recently, I'm interested in more extensive topics to achieve the goal, such as data-centric ML or human-in-loop ML.

Publications

(C: Conferences, P: Preprint, *: Equal contribution)

[C4] What Makes Better Augmentation Strategies? Augment Difficult but Not too Different

  • Jaehyung Kim, Dongyeop Kang, Sungsoo Ahn, and Jinwoo Shin

[C3] Spread Spurious Attribute: Improving Worst-group Accuracy with Spurious Attribute Estimation

[C2] Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learning [pdf][code]

[C1] M2m: Imbalanced Classification via Major-to-minor Translation [pdf][code]

[P1] Simplified Stochastic Feedforward Neural Networks [pdf]

  • Kimin Lee, Jaehyung Kim, Song Chong, and Jinwoo Shin

  • arXiv preprint 2017

Education

Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, Mar. 2019 - Present (adviser: Jinwoo Shin)

Ph.D. in Electrical Engineering

Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, Mar. 2017 - Feb. 2019 (adviser: Jinwoo Shin)

M.S. in Electrical Engineering

Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, Mar. 2012 - Feb. 2017 (Magna Cum Laude)

B.S. in Mathematics and Economics Program (Minor)

Academic Services

Conference Reviewer

  • AAAI Conference on Artificial Intelligence (AAAI): 2021, 2022

  • Neural Information Processing Systems (NeurIPS): 2021

  • International Conference on Machine Learning (ICML): 2021, 2022

  • International Conference on Learning Representations (ICLR): 2022

Journal Reviewer

  • IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

Invited Talks

Deep Learning with Imbalanced Datasets

  • Samsung Electronics Data & Information Technology Center (Oct. 2021)

Multi-aspect Analysis on Data Informativeness

  • Summer 2021 Presentation Minnesota NLP Group (Aug. 2021)

Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learning

  • NeurIPS 2020 Social ML in Korea (Dec. 2020)

Awards

  • Winner, Qualcomm Innovation Fellowship Korea, 2021

  • 3rd place, ILSVRC (ImageNet Challenge) on Object Detection, 2017