Publications
Publications
Yubin Kim, Chanwoo Park, Hyewon Jeong, Yik Siu Chan, Xuhai Xu, Daniel McDuff, Cynthia Breazeal, and Hae Won Park. Adaptive Collaboration Strategy for LLMs in Medical Decision Making. arXiv preprint arXiv:2404.15155 (2024).
Hyewon Jeong*, Nassim Oufattole*, Aparna Balagopalan, Matthew Mcdermott, Bryan Jangeesingh, Marzyeh Ghassemi & Collin Stultz (2024). Event-Based Contrastive Learning for Medical Time Series. Arxiv Preprint
Hyewon Jeong*, Nassim Oufattole*, Aparna Balagopalan, Matthew Mcdermott, Payal Chandak, Marzyeh Ghassemi & Collin Stultz (2023). Event-Based Contrastive Learning for Medical Time Series., UniReps NeurIPS Workshop.
Hyewon Jeong, Collin Stultz & Marzyeh Ghassemi (2023). Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signals. Machine Learning for Healthcare Conference. Talk
Sarah Soyeon Oh, Irene Kuang, Hyewon Jeong, Jin-Yeop Song, Boyu Ren, Jong Youn Moon, Eun-Cheol Park & Ichiro Kawachi (2023). Predicting Fetal Alcohol Spectrum Disorders Using Machine Learning Techniques: Multisite Retrospective Cohort Study. Journal of Medical Internet Research. Vol 25, e45041.
Hyewon Jeong*, Kwanhyung Lee*, Seyun Kim, Donghwa Yang, Hoon-Chul Kang, & Edward Choi (2022). Real-Time Seizure Detection using EEG: A Comprehensive Comparison of Recent Approaches under a Realistic Setting. ACM CHIL 2022. Code
* Equally Contributed
Hyewon Jeong, Siddharth Nayak, Taylor Killian, Sanjat Kanjilal & Marzyeh Ghassemi (2022). Identifying Disparities in Sepsis Treatment by Learning the Expert Policy., Reinforcement Learning for Real Life Workshop @ NeurIPS 2022 (Spotlight), Women in Machine Learning Workshop.
Hyewon Jeong, Marzyeh Ghassemi & Collin Stultz (2022). Unsupervised Deep Metric Learning for the inference of hemodynamic value with Electrocardiogram signals., Learning from Time Series for Health (Spotlight), Women in Machine Learning Workshop.
Hyewon Jeong, Kexin Yang, Ziming Wei, Yidan Ma, Intae Moon, Sanjat Kanjilal (2022). Estimating the Treatment Effect of Antibiotics Exposure on the Risk of Developing Anti-Microbial Resistance. Women in Machine Learning Workshop.
Hyewon Jeong*, Tuan Nguyen*, Eunho Yang & Sung Ju Hwang (2020). Clinical Risk Predicition with Temporal Probabilistic Asymmetric Multi-Task Learning. The Association for the Advancement of Artificial Intelligence (AAAI) 2021. Code
* Equally Contributed
Jay Heo, Junhyeon Park, Hyewon Jeong, Kwang Joon Kim, Juho Lee, Eunho Yang, Sung Ju Hwang (2020) Cost-Effective Interactive Attention Learning. International Conference on Machine Learning (ICML) 2020. Code
Hyewon Jeong*, Junseong Park*, Jin-Kyoung Shim, Jae Eun Lee, Nam Hee Kim, Hyun Sil Kim, Jong Hee Chang, Jong In Yook & Seok-Gu Kang(2019). Combined Treatment of 2'-Hydroxycinnamaldehyde and Temozolomide Suppresses Glioblastoma through Decrease in Stemness and Invasiveness. Journal of neuro-oncology, 143(1), 69-77.
* Equally Contributed
You-Hyang Song, Jae-Hyun Kim, Hye-Won Jeong, Ilsong Choi, Daun Jeong, Kwansoo Kim & Seung-Hee Lee (2017). A Neural Circuit for Auditory Dominance over Visual Perception. Neuron, 93(4), 940-954.
Methods and Devices for Pill Identification. (Patent No. 10-1943217, Patent Registration, Korea, 2017.08.)
Nguyen. A.*, Jeong, H.*, Yang, E., & Hwang, S. (2021). Clinical Risk Predicition with Temporal Probabilistic Asymmetric Multi-Task Learning. The Association for the Advancement of Artificial Intelligence (AAAI) 2021.* Equally Contributed
Although the recent multi-task learning method has been shown to be effective in improving the generalization of deep natural networks, there still exists "negative transfer problem", which results in performance degeneration in individual tasks. Existing methods try to solve this by performing loss-based knowledge transfer, but the loss cannot be used as a measure of reliability as we cannot be sure whether it results from overfitting. Here, we propose a temporal probabilistic multi-task method that performs inter-task, inter-timestep knowledge transfer based on the feature level uncertainty.