Predicting Chaos with Machine Learning
Electrical Engineering
Team 43
Rachel Larkins, Danielle Williams, Natnael Berhe, Matthew Flower
Summary
Our group has performed research into chaos theory to gain understanding of chaotic dynamics, in order to generate ground truth data for a data driven approach to modeling chaotic systems. Our data-driven approach involves the use of a type of shallow recurrent neural network called a reservoir computer. We have built a reservoir computer that can learn chaotic dynamics and can predict collapses that can be extrapolated to real world complex systems.
Our project examines chaos theory, generates ground truth data for four chaotic systems, then uses a machine learning model to learn and predict the properties of the chaotic systems from the ground truth data.
Demo Day Zoom link: https://asu.zoom.us/j/3443412705