SciML – Scientific Machine Learning
Course Description
Dive into the theory, implementation, and limitations of Scientific Machine Learning (SciML) models. In this course, you will engage in hands-on and practical activities under the expert guidance of our experienced instructors. Learning SciML will be enjoyable and rewarding as you engage in real-world applications, such as solving partial differential equations using physics-informed neural networks (PINNs) and neural operators (PyTorch implementation). By the end of this course, you’ll not only understand the theory behind SciML but also have the confidence and expertise to tackle any partial differential equation with PyTorch.
This course is one of 6 courses in the Advanced AI Techniques pilot micro-credential pathway offered by the Translational AI Center at Iowa State University.
Advanced AI Techniques Pathway Courses:
Prerequisites
- Basic Python programming
- Basic understanding of numerical methods and deep learning
Intended Audience
The course is intended for a broad audience within the spectrum of the software and technology industry, including software engineers, data scientists, data engineers, data analysts, research scientists, and software developers. The course is designed to provide a basic understanding of AI and how to use PyTorch for a broad range of audiences.
Learning Outcomes
Assessments
Course Outline
By the end of the course, you should be able to:
- Apply scientific machine learning (SciML) models to solve partial differential equations (PDEs) for engineering problems.
- Implement Physics-Informed Neural Networks (PINNs) using PyTorch to solve differential equations.
- Analyze the theory, implementation, and limitations of neural operators.
- 2 quizzes to help design and debug SciML models (there will be unlimited attempts).
- 2 coding exercise questions, which include implementing Python codes based on hands-on activities. This includes coding a simple physics-informed neural network and using state-of-the-art neural operators (FNO, DeepONet) to learn partial differential equations (PDEs) from data.
- Module 1: Introduction to Partial Differential Equations (PDEs)
- Module 2: Physics-Informed Neural Networks (PINNs) Codes
- Module 3: Operator Learning (Fourier Neural Operator)
- Module 4: Operator Learning (DeepONet)
Course Procedures
The course starts on September 6, 2024. All coursework must be completed before the course ends on October 6, 2024. The approximate time to complete this course is 16 hours. You can complete the modules at your own pace.
Live Zoom meetings will be conducted for interactive coding sessions. A suitable time for these live sessions will be determined through a group poll. The recordings of those sessions will be available soon after each meeting.
You will receive a micro-credential badge upon completing the assessments at the end.
Course Materials
Course materials are provided within the course. No additional purchase is required.
Contact Information
Contact isopd@iastate.edu for more information.
Course Developer
The Translational AI Center breaks down disciplinary silos to bring together core Iowa State artificial intelligence researchers and subject matter experts interested in applying new technologies to their work. For more information, visit Translational AI Center at Iowa State University
Registration Cost: $750.00 $500 USD (Initial Promo)
*$300 Student & Government Employee
Course Hours: 16 hours
Course Start Date: September 9, 2024
Last Day to Register: September 13, 2024
Course End Date: October 6, 2024
Course Access Time: 27 Days
*At the time of registration, you’ll be asked to create an account for this course. Use an email address ending in “.edu” or “.gov” to receive a discount. $200.00 will be immediately credit back after purchase.
Instructor
Ali Rabeh, Postdoctoral Researcher
Ali Rabeh is a postdoctoral researcher at the Translational AI Center (TrAC). He earned his PhD in Mechanical Engineering from Iowa State University. His research interests include scientific machine learning, multiphase flow of immiscible fluids, and finite element methods. He is also exploring the application of state-of-the-art neural operators for modeling fluid flow around complex geometries.
