Full Syllabus

Course Description

This course explores the integration of modern machine learning techniques with physics-based modeling and simulation. Structured around foundational principles and emerging research, we will examine differentiable simulations, physics-informed learning, reduced-order modeling, and generative AI methods for physical systems. Topics span from graph neural networks and neural operators to diffusion models and AI-driven scientific discovery. Students will engage with key readings and hands-on projects, culminating in the creation of a differentiable simulator for a chosen application.

Seminar Structure

Unlike traditional lecture-style courses, this research seminar uses a role-playing format to immerse students in the research lifecycle. Each week, we’ll read a technical paper on learning and simulation. During role-play, students will rotate through various positions. Each student will give a 5-minute presentation from their assigned perspective, followed by a class discussion.

Course Goals

• Develop differentiable simulations that combine physical principles and data-driven methods.
• Compare physics-based and data-driven modeling approaches.
• Apply graph neural networks, neural operators, and transformers to scientific problems.
• Design, manage, and communicate an open-ended research project.

Prerequisites/Corequisites

Advanced undergraduate and beginning graduate students in computer science and applied sciences.

Course Materials

Optional textbook: Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control by Steven L. Brunton and J. Nathan Kutz.

What to Expect in This Course

Students will engage in a group project throughout the semester and apply machine learning to science and engineering problems they propose. Pre-class preparation includes readings and supplementary materials. In class, frequent peer discussions, direct feedback, polling, think-pair-share, and multiple-hands-multiple-voices activities support active learning and diverse participation.

Assessment

Role play: 60% — 10 presentations, 6% each.

Final project: 40% — proposal 5%, mid-term check-in I 5%, mid-term check-in II 5%, final presentation 25%.

Final Project

Option 1: Implement and refine a differentiable simulation paper.

Option 2: Incorporate differentiable simulation into your own research.

Group size: 2–3 people.

Schedule and Communication

Weekly topics and readings: see the 2026 course schedule.

Role assignments and presentation slides will be coordinated through Slack.

Slack invitation: join the CPSC 532Z Slack workspace.

Pre-class survey: complete the survey.