Embedded Files
I am interested in building optimization frameworks and reduced-order models for complex engineering systems. My focus is on large-scale, physics-based simulations and sparse parameter identification. Passionate about delivering computationally efficient solutions that enhance model interpretability and performance.
Research Highlights
Researched and applied advanced optimization methods to extract interpretable parameters from noisy and incomplete data.
Developed and implemented sparse Hessian solvers in JAX to improve computational efficiency for large-scale problems.
Reduced Order Models for Circuits
Reduced Order Models for Circuits
Investigated reduced-order models for electrical circuits where state measurements are not directly attainable.
Applied advanced optimization algorithms to calibrate model parameters, ensuring accurate representation of circuit dynamics.
Physics Informed Neural Networks
Physics Informed Neural Networks
Engineered neural network architectures tailored to solving incompressible Stokes equations under complex boundary conditions.
Analyzed the impact of elliptic regularity theory on training stability and solution accuracy for PINNs.
Implemented efficient numerical methods to solve the Helmholtz Paraxial wave equation.
Compared and analyzed the computational complexity of spectral versus sinc-based approaches, optimizing for performance.
Publications (*in review)
*Teddy Meissner and Karl Glasner. Discovery of differential equations using sparse state and parameter regression. 2024.
*Teddy Meissner, Edward Huynh, Paul Kuberry and Pavel Bochev. A deep least-squares method for the Stokes equations. 2024.
Presentations (*upcoming)
Oral Presentations
*Teddy Meissner, Biliana Paskaleva, Pavel Bochev - ”A Sparse State-Space Model for Electrical Circuits”, Digital Twins in Engineering & Artificial Intelligence and Computational Methods in Applied Science (DTE & AICOMAS), February 2025, Paris, France.
*Teddy Meissner, Karl Glasner - ”Discovery Differential Equations Under High Data Corruption”, National Defense Science and Engineering Graduate (NDSEG) Fellow Conference, July 2025.
Poster Presentations
*Teddy Meissner, Biliana Paskaleva, Pavel Bochev - ”Sparse state-space models for circuits under ionizing radiation”, Hardened Electronics and Radiation Technology (HEART) Conference, April 2025, Monterey, CA.
Presented on My Behalf
Teddy Meissner, Edward Huynh, Paul Kuberry, Pavel Bochev - ”A deep least-squares method for the Stokes equations” presented by Pavel Bochev at the Minimum Residual & Least-Squares Finite Element Methods Conference, June 2024, Bilbao, Spain
Page updated
Google Sites
Report abuse