Introduction to Artificial Intelligence for Civil, Environmental, and Water Engineering

Catalog description

This course offers a comprehensive introduction to domain-oriented Artificial Intelligence (AI) and Machine Learning (ML) methods, equipping future engineers and scientists with cutting-edge AI/ML skills, preparing them for advanced AI tasks and making them proficient in solving complex problems with AI models. In today's rapidly evolving technological landscape, AI/ML is transforming the way problems are approached and solved across industries, including Civil and Environmental Engineering. Employers increasingly seek professionals with AI expertise to address complex challenges and drive innovation. By learning AI/ML methods in the context of their career interests, students will be well-equipped to excel in their fields and stand out in the job market. The course bridges fundamental principles of ML with domain-relevant examples taken from Civil, Environmental, and Water Engineering, providing students with the skills necessary to train and run AI/ML models in their respective fields. Examples include examining pavement performance with a Random Forest, predicting water quality and floods with Recurrent Neural Networks, and identifying microbes and examining building safety with Convolutional Neural Networks. The covered methods are selected to cover domain problems and practices, so that by the end of the course, students will be familiar with AI model concepts and the usefulness of AI and ML models in their domains and industries. They will be exposed to hands-on coding exercises, model training, benchmarking, and finetuning. The content related to statistics and math is limited. For those unfamiliar with Python, the first few sessions of the course will include exercises on Python and PyTorch, a popular ML library in Python. AI algorithms to be covered include regularized linear regression, regression trees and random forest, boosting, convolutional and recurrent neural networks, unsupervised learning, generative models, and the effective use of generative AIs for model training, all in the context of Civil, Environmental, and Water Engineering (CEWE) applications.