Woojin

AI researcher

Author Image

About

I am currently a Senior AI Researcher at TelePIX, serving as Technical Research Personnel under Korea’s alternative military service program. I received my M.S. in Artificial Intelligence from Yonsei University, where I was advised by Prof. Noseong Park (now at KAIST).

My research focuses on scientific machine learning (SciML) and AI for satellite applications, spanning Earth observation and geospatial data analysis. Within this scope, I am particularly interested in implicit neural representations, neural operators, and foundation models for physical systems, especially for modeling high-dimensional scientific processes and spatiotemporal satellite observations. More broadly, I aim to develop numerically grounded and interpretable learning methods that bridge numerical analysis and modern machine learning.

I enjoy collaborating with researchers who share similar interests. If you are interested in my research or would like to collaborate, please feel free to contact me at woojin.py@gmail.com.

Publication

Conference & Journal

Platonic Task Arithmetic
J Park, W Cho†
(Under review)

Escaping Spectral Bias without Backpropagation: Fast Implicit Neural Representations with Extreme Learning Machines
W Cho*, J Park*
(Under review)
Paper

Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data
W Cho, K Lee, N Park, D Rim, G Welper
(Under review)
Paper

ASTRA: A Large-Scale Multi-Satellite Benchmark for Onboard 6-DoF Spacecraft Pose Estimation
W Cho*, J Park*, S Immanuel, J Park, S Chin, J Wang
(Under review)

DRIFT: Dynamics-aware Robust Inference with Latent Filtering for Temporal Satellite Pose Estimation
J Park*, W Cho*
(Under review)

MaD-Scientist: AI-based Scientist solving Parabolic PDEs using Massive Prior Data
M Kang, D Lee, W Cho, J Park, K Lee, A Gruber, Y Hong, N Park
(Under review)
Paper

Meta-learning Structure-Preserving Dynamics
C Jing, U Mudiyanselage, W Cho, M Jo, A Gruber, K Lee
ICML 2026
Paper

Summarize First, Download Later: Onboard VLMs for Bandwidth-Efficient Earth Observation
J Park, S Sim, W Cho, D Kwon
IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2026 (Oral)
Paper

ELMZip: Onboard Satellite Image Compression via Extreme Learning Machines for Efficient Downlink
W Cho, J Park, S Sim, S Immanuel, J Heo, D Kwon
IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2026
Paper

Basis-Oriented Low-rank Transfer for Few-Shot and Test-Time Adaptation
J Park, W Cho, J Heo, D Kwon, K Lee
CVPR 2026
Paper

Fourier-Modulated Implicit Neural Representation for Multispectral Satellite Image Compression
W Cho*, S Immanuel*, J Heo, D Kwon
IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2025 (Oral)
Paper Code

PDEfuncta: Spectrally-Aware Neural Representation for PDE Solution Modeling
M Jo*, W Cho*, U Mudiyanselage, S Lee, N Park, K Lee
NeurIPS 2025
Paper Code

FastLRNR and Sparse Physics Informed Backpropagation
W Cho, K Lee, N Park, D Rim, G Welper
Results in Applied Mathematics 2025
Paper

Neural Functions for Learning Periodic Signal
W Cho*, M Jo*, K Lee, N Park
ICLR 2025
Paper Code

Promoting Sparsity In Continuous-Time Models To Learn Delayed Dependence Structures
F Wu, W Cho, D Korotky, S Hong, D Rim, N Park, K Lee
CIKM 2024
Paper

Parameterized Physics-informed Neural Networks for Solving Parameterized PDEs
W Cho, M Jo, H Lim, K Lee, D Lee, S Hong, N Park
ICML 2024 (Oral, Top 1.52%)
Paper Code Presentation

Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformer
Y Yu, J Choi, W Cho, K Lee, N Kim, K Chang, C Woo, I Kim, S Lee, J Yang, S Yoon, N Park
ICLR 2024
Paper Code

Operator-learning-inspired Modeling of Neural Ordinary Differential Equations
W Cho*, S Cho*, H Jin, J Jeon, K Lee, S Hong, D Lee, J Choi, N Park
AAAI 2024
Paper Code

Hypernetwork-based Meta-Learning for Low-Rank Physics-Informed Neural Networks
W Cho, K Lee, D Rim, N Park
NeurIPS 2023 (Spotlight, Top 3.06%)
Paper Code

Workshop

On-Orbit Demonstration of Backpropagation-Free Neural Compression for Earth Observation
W Cho, J Park, S Sim, D Kwon
(Under review)

FLAME: Physics-Guided Neural Operators for Onboard Satellite Methane Detection in Hyperspectral Imagery
J Heo, J Park, S Sim, B Choi, W Cho†
ICML Workshop 2026
Paper

Sequential Dataset for Satellite Pose Estimation and a Frequency-Space Neural Operator for HIL-Free Generalization Benchmarking
W Cho*, J Park*, S Immanuel, S Chin, J Wang
CVPR Workshop 2026
Paper

Onboard Latent Kalman Filtering for Robust Spacecraft Pose Estimation
J Park*, W Cho*
CVPR Workshop 2026
Paper

PIANO: Physics-informed Dual Neural Operator for Precipitation Nowcasting
S Chin, J Park, W Cho†
NeurIPS Workshop 2025
Paper

Unveiling the Potential of Superexpressive Neural Networks in Implicit Neural Representations
U Mudiyanselage, W Cho, M Jo, N Park, K Lee
ICLR Workshop 2025
Paper

Tackling Few-Shot Segmentation in Remote Sensing via Inpainting Diffusion Model
S Immanuel, W Cho, J Heo, D Kwon
ICLR Workshop 2025 (Best Paper)
Paper Code

Can we pre-train ICL-based SFMs for the zero-shot inference of the 1D CDR problem with noisy data?
M Kang, D Lee, W Cho, K Lee, A Gruber, N Trask, Y Hong, N Park
NeurIPS Workshop 2024
Paper

Neural Compression for Multispectral Satellite Images
W Cho*, S Immanuel*, J Heo, D Kwon
NeurIPS Workshop 2024
Paper Code Project

Extension of Physics-informed Neural Networks to Solving Parameterized PDEs
W Cho, M Jo, H Lim, K Lee, D Lee, S Hong, N Park
ICLR Workshop 2024
Paper Code

Education

  • Yonsei University (Aug. 2024)

    M.S in Artificial Intelligence

  • Yonsei University (Aug. 2022)

    B.S in Atmospheric science
    B.S in Electrical electronic engineering

  • Sejong Science High School (Feb. 2017)

Career

  • TelePIX ( Jun. 2024 - Present )

    Senior AI Researcher (Space AI team)

  • Arizona State University ( Jan. 2024 - Jun. 2024 )

    Visiting Researcher (hosted by Prof.Kookjin Lee)

Academic Activities

Conference Reviewer

  • Conference on Neural Information Processing Systems (NeurIPS)
  • International Conference on Machine Learning (ICML)
  • International Conference on Learning Representations (ICLR)
  • Association for the Advancement of Artificial Intelligence (AAAI)
  • Conference on Computer Vision and Pattern Recognition (CVPR)

ESA-NASA International Workshop on AI Foundation Model for Earth Observation

  • A Unified Framework for Multi-resolution and Multi-spectral Satellite Imagery in Foundation Model Training (Lead author)
  • Multi-modal Foundation Model for EO and SAR Images (Lead author)

AGU Fall Meeting 2025

Invited Talk

  • AI for Science: Learning Scientific Signals via Implicit Neural Representations (hosted by Solver X)
  • Physics-informed Machine Learning and the Road to Scientific Foundation Models (hosted by ETRI)
  • AI for Computational Science and Space Exploration (hosted by Postech, EFFL)
  • Scientific Machine Learning (hosted by KIAS: Korea Institute For Advanced Study)
  • Parameterized Physics-informed Neural Networks for Parameterized PDEs (hosted by ML2)
  • Latest Trends in Machine Learning based Physics Simulation (hosted by Samsung Electronics)
  • Physics-informed Neural Networks for Solving PDEs (hosted by Alsemy)

Scholarship