I work on Bayesian optimization and like to see it as a framework for decision-making under uncertainty. I'm also into using Bayesian inference in different scientific domains.
Before Cornell, I did my undergrad and masters at Korea University in Seoul, South Korea, advised by
Seungjun Baek.
Email: sihwapark[at]cs[dot]cornell[dot]edu
News
Sep 24, 2026
A new paper got accepted to NeurIPS 2026!
Projects
Bayesian Optimization for Alpha-particle Confinement in Stellarators
Uses constrained Bayesian optimization to search for stellarator magnetic
configurations that better confine alpha particles, directly optimizing a
confinement measure computed from GPU-accelerated particle-tracing simulation.
Thompson Sampling using Prior-fitted Diffusion Transformers
Uses diffusion transformers that perform Thompson sampling in-context at
inference time after pre-training on functions drawn from a prior.
Introduces practical non-Gaussian priors that can overcome limitations of Gaussian processes.
Cost-aware Multi-objective Bayesian Optimization via Gittins Indices
Generalizes Pandora’s Box Gittins Index framework to multi-objective and
cost-aware settings, studies the optimality of index-based policies in
multi-objective decision problems, and proposed Gittins index–based multi-objective methods.
Publications
Thompson Sampling using Prior-fitted Diffusion Transformers S. Park*, J. Zhu*, V. Jain, S.-Y. Chou, A. Terenin NeurIPS 2026
NeBLa: Neural Beer-Lambert for 3D Reconstruction of Oral Structures from Panoramic Radiographs S. Park, S. Kim, D. Kwon, Y. Jang, I.-S. Song, S. Baek AAAI 2024 arXiv
3D Teeth Reconstruction from Panoramic Radiographs using Neural Implicit Functions S. Park, S. Kim, I.-S. Song, S. Baek MICCAI 2023 ยท top 14% arXiv