Enhancing Differential Equation Modeling with Large Language Models
Day 3 | 5:00 PM–5:30 PM
"Enhancing Differential Equation Modeling with Large Language Models"
Presented by:
Marcelo Guerra Hahn, Lake Washington Institute of Technology, Kirkland WA USA
https://qubeshub.org/community/groups/simiode/expo/2025
Abstract: Differential equations are crucial for predicting dynamic systems in science and engineering. Developing and solving these equations is challenging, requiring both math expertise and computational skills. This presentation shows how large language models (LLMs) can simplify and improve the modeling process, making it more accessible and efficient. We will demonstrate how to solve an exponential population growth problem with the help of an LLM. Starting from the problem description, the LLM helps identify key variables and work on designing the differential equation. The LLM can also guide the students through solving the equation and visualizing population changes over time. The LLM can also expand on growth rate, initial conditions, and model predictions. Integrating large language models (LLMs) into the workflow allows educators and students to concentrate on interpreting results and addressing practical problems while getting support when dealing with technical details. This presentation then demonstrates the potential of LLMs as collaborative tools in mathematical modeling, enabling users of varying expertise levels to engage effectively with differential equations.