Teaching

2024 Spring

The course provides a basic understanding of how to make use of machine learning techniques for dynamical system. For this purpose, the class covers an introduction to dynamical system theory and how to model and control systems on the basis of observed data. The main topics for dynamical systems include the representation of nonlinear systems, stability analysis, basic control concepts and identification methods. The main machine learning approaches that will be discussed are supervised learning techniques with focus on neural networks, bayesian approaches and physics-informed learning. Typical applications of the approaches presented in this course focus on mechanical and robotic systems.

Materials (Requires Vanderbilt credentials)

2023 Fall

Theoretical and algorithmic foundations of supervised learning, unsupervised learning, and reinforcement learning. Linear and nonlinear regression, kernel methods, support vector machines, neural networks and deep learning methods, instance-based methods, ensemble classifiers, clustering and dimensionality reduction, value and policy iteration. Explainable AI, ethics, and data privacy.

Materials (Requires Vanderbilt credentials)

2023 Spring

The course provides a basic understanding of how to make use of machine learning techniques for dynamical system. For this purpose, the class covers an introduction to dynamical system theory and how to model and control systems on the basis of observed data. The main topics for dynamical systems include the representation of nonlinear systems, stability analysis, basic control concepts and identification methods. The main machine learning approaches that will be discussed are supervised learning techniques with focus on neural networks, bayesian approaches and physics-informed learning. Typical applications of the approaches presented in this course focus on mechanical and robotic systems.

Materials (Requires Vanderbilt credentials)