Incorporating Modeling with Neural Networks into Undergraduate Differential Equations Courses

Channel:
Subscribers:
439
Published on ● Video Link: https://www.youtube.com/watch?v=LCOrx7SicgU



Duration: 0:00
3 views
0


Day 1 | 4:30 PM–5:00 PM

"Incorporating Modeling with Neural Networks into Undergraduate Differential Equations Courses"

Presented by:
Abhinav Chand and Andrew Bennett, Kansas State University, Manhattan KS USA

https://qubeshub.org/community/groups/simiode/expo/2025

Abstract: With the goal of including more hands-on learning and machine learning modules into our undergraduate differential equations course, we performed experiments with the spring mass system, used the Physlets Tracker App to collect data, validated the textbook model (linear model). and created a neural network to learn the nonlinearity of damping and improve upon the linear model. In our proposed assignment module, students will learn the process of validating textbook models, conducting experiments, collecting data, and using neural networks to produce nonlinear variations on dynamical systems, as well as the limitations of machine learning.