Baek is an applied geometer, scientist, and entrepreneur. He studies the space of shapes using machine learning. Baek’s educational background is in mechanical and aerospace engineering.
Prior to joining UVA in 2021, he was an Assistant Professor at the University of Iowa, where he taught courses on deep learning. There, he also founded and directed the Visual Intelligence Laboratory, which conducts fundamental research in computational geometry, vision, and machine learning. Baek’s research interests include geometric data analysis, geometric deep learning, scientific machine learning, and data-driven design.
Baek’s published research is extensive, including “Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materials” and “Deep segmentation networks predict survival of non-small cell lung cancer.”
Baek holds a PH.D. in Mechanical and Aerospace Engineering from Seoul National University and a B.S. in Mechanical and Aerospace Engineering from Seoul National University.
PH.D., Mechanical and Aerospace Engineering, Seoul National University
B.S., Mechanical and Aerospace Engineering, Seoul National University
Load forecasting is critical for effective scheduling and operation of power systems, which are becoming increasingly complex and uncertain, especially with the penetration of distributed power. This paper proposes a data-driven deep learning framework to forecast the short-term power load.
IEEE Transactions on Emerging Topics in Computational Intelligence / March 2019 / Coauthors: Tinghui Ouyang, Yusen He, Huajin Li, Zhiyu Sun, Stephen Baek
In this paper, we propose an improved gesture control interface for 3D modeling manipulation tasks that possesses conventional interface level usability with low user fatigue while maintaining a high level of intuitiveness.
Computer-Aided Design / January 2014 / Coauthors: Junbong Song, Sungmin Cho, Stephen Baek, Kunwoo Lee, Hyunwoo Bang
Additive Manufacturing (AM) increases much design freedom for designers to conceive complex parts. However, the increased complexity makes the manufacturability analysis difficult for the designed parts when applying traditional methods. To solve this problem, this paper introduces a new feature-based method for manufacturability analysis in AM by using Heat Kernel Signature.
Computer-Aided Design and Applications / November 2018 / Coauthors: Yang Shi, Yicha Zhang, Stephen Baek, Wout De Backer, Ramy Harik
Get the latest news
Subscribe to receive updates from the School of Data Science.