Numerical Integration of Chaotic Dynamics: Uncertainty Propagation & Vectorized Integration @Eigensteve

Uploaded December 2022 | Updated September 2026, 2 weeks ago

This video introduces the idea of chaos, or sensitive dependence on initial conditions, and the importance of integrating a bundle of trajectories to propagate uncertainty. We also explore how to vectorize numerical integration in Python and Matlab to make the algorithm orders of magnitude more efficient.

Playlist: youtube.com/playlist?list=PLMrJAkhIeNNTYaOnVI3QpH7jgULnAmvPA
Course Website: http://faculty.washington.edu/sbrunton/me564/

@eigensteve on Twitter
eigensteve.com
databookuw.com

This video was produced at the University of Washington

%%% CHAPTERS %%%
0:00 Propagating uncertainty with bundle of trajectory
6:30 Slow Matlab code example
11:06 Fast Matlab code example
17:09 Python code example

Numerical Integration of Chaotic Dynamics: Uncertainty Propagation & Vectorized Integration

Deriving Forward Euler and Backward/Implicit Euler Integration Schemes for Differential Equations

New Advances in Artificial Intelligence and Machine Learning

Deep Reinforcement Learning for Fluid Dynamics and Control

The Connection Between the Exponential Distribution and the Poisson Process

Population Statistics and Random Sampling

The Heat Equation and the Steady State Heat Distribution via Laplaces Equation

Numerical Simulation of Ordinary Differential Equations: Integrating ODEs

Introduction to Statistics and Data Analysis