You Rim Choi
I am a Ph.D. student at the Graduate School of Data Science, Seoul National University, advised by Prof. Hyung-Sin Kim.
I develop machine learning methods and systems that translate into real-world impact across domains, with a particular interest in AI+X. My current focus is AI for healthcare, particularly Sleep AI, using video, images, and multimodal time-series data. I also work on enabling methods for practical deployment, including on-device AI, semi-supervised learning, and federated learning, as well as sensor-control–aware AI systems for physical AI applications.
Previously, I earned my B.S. from POSTECH and my M.S. from Seoul National University in Materials Science and Engineering, where I conducted research on nanoelectronic materials under the supervision of Prof. Ho Won Jang. This background continues to inspire my interest in applying AI to broader science and engineering problems.
news
| May 23, 2026 | Our paper on Video-based OSA detection has been accepted for Nature Communications |
|---|---|
| May 01, 2026 | Our paper on Sleep Foundation Model has been accepted for ICML 2026 |
| Apr 10, 2026 | Our paper on Bio-Inspired AI has been accepted for MobiSys 2026 |
| Nov 08, 2025 | Our paper on Open-set Semi-supervised learning has been accepted for AAAI 2026 |
| Nov 07, 2025 | Our work on Sleep AI has won Best Poster Runner-up Award at MobiCom 2025 |
| Sep 01, 2025 | I have been selected for the NRF PhD Fellowship |
| Dec 10, 2024 | Our paper on Sleep AI has been accepted for npj Digital Medicine (IF 15.1, Top 4% in Medical Informatics) |
| Oct 04, 2024 | Our paper on Federated learning has been accepted for SenSys 2024 |
| Jun 07, 2024 | Our work on Sleep AI has won Best Poster Award at MobiSys 2024 |
selected publications
- JournalNon-contact On-device Detection of Obstructive Sleep Apnea from Infrared VideoNature Communications, 2026Impact Factor: 15.7
- Conference[Emerging Ideas] Artificial Tripartite Intelligence: A Bio-Inspired, Sensor-First Architecture for Physical AIIn ACM MobiSys, 2026Acceptance ratio: 22.4%
- ConferenceLet the Void Be Void: Robust Open-Set Semi-Supervised Learning via Selective Non-AlignmentIn AAAI, 2026Acceptance ratio: 17.6%
- JournalExplainable vision transformer for automatic visual sleep staging on multimodal PSG signalsnpj Digital Medicine, 2025Impact Factor (2024): 15.1