Jaehwan Jeon

Jaehwan Jeon

Jaehwan Jeon is a Postdoctoral Fellow in Civil and Mineral Engineering at the University of Toronto. His research tackles the two bottlenecks of structural safety prediction — computation and expertise — scaling both with AI while making that AI trustworthy.

Publications

You can also find my articles on my Google Scholar profile.

Ph.D. Dissertation

  1. Jeon, J. (2026). Prediction and Uncertainty Quantification of Nonlinear Dynamic Response of Structures via Physics-informed Deep Learning. Doctoral dissertation, Seoul National University.

Journal Articles

  1. Joo, J., J. Jeon, and J. Song (2026). Physics-based Encoder-only Transformer for Structural Hysteresis Prediction and Dynamic Analysis. Earthquake Engineering & Structural Dynamics. — In press

  2. Guo, F., J. Song, Y. Liu, J. Joo, and J. Jeon (2026). A Unified Rate-Dependent Rheological Model for High Damping Rubber Bearings: Seismic Performance Assessment in Cold Climates. Engineering Structures. Vol. 367, 123564.

  3. Jeon, J., J. Song, and O. Kwon (2026). Ensemble-based Uncertainty Quantification and Decomposition of Probabilistic Surrogate Models Using Bayesian Neural Networks. Structural Safety. Vol. 121, 102698. — Ranked as most downloaded paper

    This study presents a new framework for decomposing predictive uncertainty in ensembles of Bayesian neural networks, systematically separating irreducible randomness from reducible model-form and data-induced uncertainties.

  4. Jeon, J., O. Kwon, and J. Song (2026). Unified Hysteresis Modeling via Physics-based Deep Learning and Data Augmentation. Earthquake Engineering & Structural Dynamics. Vol. 55, No. 2, 379-396.

    This paper proposes a unified hysteresis modeling framework based on deep learning, leveraging physics-encoded architecture and physics-informed loss functions to capture complex nonlinearities under stochastic excitations.

  5. Jeon, J., and J. Song (2025). Neural Network–Augmented Physics Models Using Modal Truncation for Dynamic MDOF Systems under Response-Dependent Forces. Journal of Engineering Mechanics. Vol. 151, No. 1. — Ranked as most read & most cited paper

    This paper develops a neural network-augmented physics model for MDOF systems under response-dependent forces using modal truncation.

Under Review

  1. Jeon, J., C. H. Lee, and T. Kim. Automating Structural Reliability Analysis with a Multi-Agent Large Language Model Framework. (Under review).

Conference Papers

  1. Park, S., J. Jeon, and J. Song (2026), Probabilistic SINDy Surrogate Model for Ensembled Uncertainty-Quantification of Hysteretic Structures, 17th International Workshop on Advanced Smart Materials and Smart Structures Technology (ANCRiSST 2026), July 24-25, Turin, Italy.

    (Abstract)

  2. Park, S., J. Jeon, and J. Song (2026), Ensemble-based Uncertainty Quantification for Probabilistic Surrogate Models using SINDy, 10th International Symposium on Reliability Engineering and Risk Management (ISRERM 2026), June 28-July 1, Sapporo, Japan.

  3. Joo, J., J. Jeon, and J. Song (2026), Uncertainty Quantification in Unified Hysteresis Modeling Using a Partially Bayesian Neural Architecture, 10th International Symposium on Reliability Engineering and Risk Management (ISRERM 2026), June 28-July 1, Sapporo, Japan.

  4. Joo, J., J. Jeon, and J. Song (2026), Uncertainty Quantification in Unified Hysteresis Modeling Using a Partially Bayesian Neural Architecture, Engineering Mechanics Institute Conference 2026 (EMI 2026), June 2-5, Boulder, USA. — Selected as EMI 2026 PMC student competition finalist

    (Extended Abstract)

  5. Park, S., J. Jeon, and J. Song (2026), Ensemble-based Uncertainty Quantification for Probabilistic Surrogate Models using SINDy, Engineering Mechanics Institute Conference 2026 (EMI 2026), June 2-5, Boulder, USA.

    (Extended Abstract)

  6. Joo, J., J. Jeon, and J. Song (2026), Multi-Modal Deep Learning-Based Seismic Damage Identification for Nuclear Power Plant Structures, Transactions of the Korean Nuclear Society Spring Meeting, May 7-8, Jeju, Korea. — KNS Spring Meeting Best Paper Award

  7. Jeon, J., J. Song, and O. Kwon (2025), Uncertainty Quantification of Ensemble Prediction for Structural Hysteretic Behavior using Bayesian Neural Networks, Computational Structural Engineering Institute of Korea Annual Conference, November 28-29, Daejeon, Korea. — COSEIK Best Paper Presentation Award

  8. Joo, J., J. Jeon, and J. Song (2025), Dynamic Analysis of Hysteretic Structures Using Transformer Models with Transfer Learning, Civil Engineering Conference in the Asian Region, October 21-24, Jeju, Korea.

  9. Jeon, J., J. Song, and O. Kwon (2025), Enhancing Uncertainty Quantification in Civil Engineering through Bayesian Neural Network Ensembles, Civil Engineering Conference in the Asian Region, October 21-24, Jeju, Korea.

  10. Joo, J., J. Jeon, and J. Song (2025), Development of a Transformer-based Unified Hysteretic Behavior Model for Seismic Response Prediction of Nuclear Power Plant Structures, 2025 Korea Electric Power Industry Code-Week, August 11-14, Busan, Korea.

  11. Joo, J., J. Jeon, and J. Song (2025), Physics-Based Transformer Model for Predicting Nonlinear Hysteresis in Structures, 14th International Conference on Structural Safety and Reliability (ICOSSAR'25), June 1-6, Los Angeles, USA.

  12. Jeon, J., O. Kwon, and J. Song (2025), Uncertainty Quantification of Hysteresis Prediction in Structural Dynamics Using Bayesian Neural Networks, 14th International Conference on Structural Safety and Reliability (ICOSSAR'25), June 1-6, Los Angeles, USA. — Selected as ICOSSAR 2025 student competition finalist

    (Extended Abstract)

  13. Jeon, J., J. Song, and O. Kwon (2025), Bayesian Neural Network Ensembles for Quantifying Uncertainty in Deep Learning-based Surrogate Models, Engineering Mechanics Institute Conference 2025, May 27-30, Anaheim, USA.

    (Extended Abstract)

  14. Jeon, J., O. Kwon, and J. Song (2024), Probabilistic Seismic Hysteresis Modeling Method Using Bayesian Neural Networks, Korean Society of Civil Engineering 2024 Convention, October 16-18, Jeju, Korea. — KSCE Convention Best Paper Award

  15. Jeon, J., O. Kwon, and J. Song (2024), Uncertainty Quantification of Deep Learning Hysteresis Models Using Bayesian Neural Network with Concrete Dropout, 8th Asia-Pacific Symposium on Structural Reliability and Its Applications, September 29-October 2, Ulsan, Korea.

  16. Jeon, J., O. Kwon, and J. Song (2024), Long Short-term Memory Model for Dynamic Analysis of Hysteretic Structures Subjected to Earthquakes, 18th World Conference on Earthquake Engineering, June 30-July 5, Milan, Italy.

    (Extended Abstract)

  17. Jeon, J., J. Song, and O. Kwon (2024), Model Uncertainty Quantification and Selection for Deep Learning-based Simulation of Hysteresis with Stiffness and Strength Degradations, 9th European Congress on Computational Methods in Applied Sciences and Engineering, June 3-7, Lisbon, Portugal.

  18. Jeon, J., J. Song, and O. Kwon (2024), Integrating Physics-Based Deep Learning Models and Data Augmentation to Reduce Uncertainties in Hysteresis Modeling, Engineering Mechanics Institute Conference and Probabilistic Mechanics & Reliability Conference 2024 (EMI/PMC 2024), May 28-31, Chicago, USA. — Selected as EMI/PMC 2024 student competition finalist

    (Extended Abstract)

  19. Jeon, J., O. Kwon, and J. Song (2024), Physics-informed Deep Learning Model for Prediction of Complex Hysteresis, 27th International Conference on Structural Mechanics in Reactor Technology, March 3-8, Yokohama, Japan.

    (Extended Abstract)

  20. Jeon, J., and J. Song (2023), Generalized Seismic Hysteresis Modeling Method Using Long Short-Term Memory Model, Korean Society of Civil Engineering 2023 Convention, October 18-20, Yeosu, Korea.

  21. Jeon, J., and J. Song (2023), Deep-Learning-Augmented Physics Models to Predict Nonlinear Dynamic Responses of Multi-Degree-of-Freedom Structures, 14th International Conference on Applications of Statistics and Probability in Civil Engineering, July 9-13, Dublin, Ireland.

    (Extended Abstract)

  22. Jeon, J., and J. Song (2023), Dynamic Response Prediction of Nonlinear MDOF Systems by Neural-Network-Augmented Physics Models, Engineering Mechanics Institute Conference 2023, June 6-9, Atlanta, USA.

    (Extended Abstract)

  23. Jeon, J., and J. Song (2022), Physics-Informed Neural Network for Dynamic Responses of Structures Based on Monitoring Data, The 2022 World Congress on Advances in Civil, Environmental, & Materials Research, August 16-19, Seoul, Korea.

  24. Jeon, J., and J. Song (2022), Prediction of Dynamic Responses of Structures Using Neural Network Augmented Physics Model, Computational Structural Engineering Institute of Korea 2022 Annual Conference, April 13-15, Jeju, Korea.

  25. Jeon, J., and J. Song (2021), Development of Artificial Neural Network to Estimate Nonlinear Damping Force Under Wind Loadings Using Acceleration and Force Data - Focusing on Small Damping Force Identification, Korean Institute of Bridge and Structural Engineers 2021 Conference: Digital Twin Technology in Infrastructure, November 26, Virtual conference.

  26. Jeon, J., and J. Song (2021), Development of Artificial Neural Network to Estimate Damping Force Under Wind Loadings Using Acceleration and Force Data, Korean Society of Civil Engineers 2021 Convention, October 20-22, Gwangju, Korea.