I am Juyeop Han, a Ph.D. candidate at MIT LIDS and a member of AERA Lab, advised by Prof. Sertac Karaman. My research focuses on building reliable and trustworthy machine learning for autonomous systems, with an emphasis on understanding when and why models fail and how to leverage this information for safe and generalizable autonomy.
Research Interests
- Uncertainty Quantification (UQ) — Developing principled methods to quantify predictive uncertainty in deep neural networks and generative models, enabling systems to know what they don’t know.
- Reliable and Generalizable Robot Learning — Leveraging uncertainty-aware models to build generalizable and robust perception and decision-making systems for autonomous robots operating in unstructured environments.
I received my MS degree in Aerospace Engineering at KAIST advised by Prof. Han-Lim Choi and BS degree in Mechanical Engineering (Summa Cum Laude) at Seoul National University (SNU) under the supervision of Prof. Sung-hoon Ahn. I worked as a visiting scholar at University of Colorado Boulder (CU Boulder), where I collaborated with Prof. Zachary Sunberg between October 2022 and February 2023.
News
Jan 2025 Our paper on CaLiSa-NeRF has been published at WACV 2025 GeoCV Workshop.
Jul 2024 I have been selected to receive a 2024–2025 MathWorks Mechanical Engineering Fellowship.
Jun 2024 My paper on NVINS has been accepted for oral presentation at IROS 2024.
Sep 2023 I have joined AERA Lab at MIT as a graduate research assistant!
Jul 2023 Our RSS IDMAV workshop was successfully held.
Mar 2023 Our proposal on RSS workshop IDMAV has been accepted.
Jan 2023 My paper on Computing Forward Reachable Sets for Nonlinear Adaptive Multirotor Controllers has been accepted at ACC 2023.
Jan 2023 My paper on DS-K3DOM has been accepted at ICRA 2023.