Papers
1
Total Citations
6
H-Index
1
About
Jisu Kang is a researcher whose work focuses on advancing machine learning techniques for handling incomplete and imperfect real-world data. Their primary research areas include time series analysis, missing data imputation, and self-supervised learning. Kang’s most notable contribution is the development of **RDIS (Random Drop Imputation with Self-Training)**, a novel framework for imputing missing values in time series data without requiring ground truth labels. This work, published in 2020, addresses a critical challenge in fields like healthcare, meteorology, and robotics, where data gaps are common. By combining random drop strategies with self-training, RDIS enables models to learn robust imputation patterns implicitly, achieving strong performance even with high missing rates. With 6 citations, this paper has already influenced subsequent research on self-supervised learning for time series. Kang’s work stands out for its practical impact—bridging the gap between theoretical imputation methods and real-world deployment. Their approach offers a scalable, label-free solution that is particularly valuable in domains where labeled data is scarce or expensive. As a rising voice in data-centric AI, Jisu Kang continues to push the boundaries of how machines can learn from incomplete information.
Research Focus
Key Achievements
Top Papers
- 1