PhD student at Carnegie Mellon University working on simulation, SciML, AI for Science, FEM, and machine learning.
Currently in the Computational Bio-Modeling Lab advised by Prof. Yongjie Jessica Zhang.

I am a second-year PhD student at Carnegie Mellon University in Prof. Yongjie Jessica Zhang's Computational Bio-Modeling Lab. My research interests sit at the intersection of scientific computing, physics-based simulation, and machine learning.

I care about building computational methods that make simulation pipelines faster, more reliable, and easier to use for real scientific and engineering problems.

• Research areas: simulation, scientific machine learning, AI for Science, FEM, and ML
• Current focus: learning-enhanced simulation and automated simulation workflows
• Application interests: cardiovascular flow, biomedical simulation, and computational engineering
• Lab: Computational Bio-Modeling Lab, CMU

Developing and exploring SciML methods that combine physical structure, numerical simulation, and data-driven models.

Working on simulation-centered workflows for cardiovascular flow problems, with interest in computational efficiency, model reliability, and biomedical relevance.

Building agent-based tools that help automate simulation setup, execution, analysis, and iteration.

NeuronTransportGALDS contains the official code for GALDS, a graph-autoencoder-based latent dynamics surrogate model for predicting neurite material transport.

I have also worked on PINN-related topics and physics-informed learning for simulation problems.

• GALDS: A Graph-Autoencoder-based Latent Dynamics Surrogate model to predict neurite material transport
  Tsung Yeh Hsieh, Yongjie Jessica Zhang.
  Code · arXiv · Google Scholar

• A multiscale stabilized physics informed neural networks with weakly imposed boundary conditions transfer learning method for modeling advection dominated flow
  Tsung Yeh Hsieh, Tsung-Hui Huang. Engineering with Computers, 2024.
  DOI · Google Scholar

• Scientific machine learning and AI for Science
• Finite element methods and physics-based simulation
• Machine learning for PDEs and dynamical systems
• Biomedical and cardiovascular flow simulation
• Surrogate modeling and reduced-order modeling
• Agentic workflows for scientific computing

I am always open to thoughtful conversations about simulation, scientific machine learning, and computational engineering.

Building learning-enhanced simulation tools for science and engineering.