This course covers fundamental machine learning concepts and hands-on experience about smart infrastructure system applications. Lectures will cover the “full stack” of machine learning, including data cleaning and pre-processing, machine learning, and reinforcement learning. The covered topics include: collecting and processing real-world project data, detecting abnormal data, and imputing missing data; analyzing the data with a variety of machine learning methods including linear & logistic regression, decision tree, SVM, unsupervised learning (clustering, PCA), and advanced machine learning methods (ensemble learning and deep learning). Last, we will introduce reinforcement learning to build autonomous infrastructure systems that learn to make good decisions. One key feature of this course is its application-driven nature. All applications of this course are about smart infrastructure systems. Through those applications, students will know how ML can be applied in their future career and research.