Jaeyoung Park

Jaeyoung Park

I am a PhD candidate at KAIST, advised by Prof. Steven Whang. My research interests are in the area of machine learning and data management.

Posts by Collection

portfolio

publications

iFlipper: Label Flipping for Individual Fairness

Published in SIGMOD, 2023

iFlipper improves individual fairness by minimally flipping inconsistent labels through an LP-based algorithm with theoretical guarantees.

Recommended citation: Hantian Zhang, Ki Hyun Tae, Jaeyoung Park, Xu Chu, and Steven Euijong Whang. 2023. IFlipper: Label Flipping for Individual Fairness. Proc. ACM Manag. Data 1, 1, Article 8 (May 2023), 26 pages. https://doi.org/10.1145/3588688.
Download Paper | Download Bibtex

Falcon: Fair Active Learning Using Multi-Armed Bandits

Published in VLDB, 2024

Falcon is a scalable fair active learning framework that improves model fairness through strategic sample selection.

Recommended citation: Ki Hyun Tae, Hantian Zhang, Jaeyoung Park, Kexin Rong, and Steven Euijong Whang. 2024. Falcon: Fair Active Learning Using Multi-Armed Bandits. Proc. VLDB Endow. 17, 5 (January 2024), 952–965. https://doi.org/10.14778/3641204.3641207.
Download Paper | Download Slides | Download Bibtex

Fair Class-Incremental Learning using Sample Weighting

Published in SIGKDD, 2026

FSW is a fair class-incremental learning framework that mitigates unfair catastrophic forgetting by adjusting sample weights to balance accuracy and group fairness.

Recommended citation: Jaeyoung Park, Minsu Kim, and Steven Euijong Whang. 2026. Fair Class-Incremental Learning using Sample Weighting. arXiv preprint arXiv:2410.01324.
Download Paper | Download Bibtex

talks

teaching