CMU launches 10-749 AI for Scientific Computing: PINNs, neural operators, climate diffusion models

risteski_a · x · 2026-09-04

CMU professor Andrej Risteski is teaching a new Fall 2026 course, 10-749 "AI for Scientific Computing," on learned surrogates for scientific computing applied to PDE solving, forecasting, and sampling.

The syllabus spans physics-informed neural networks (PINNs), neural operators and PDE surrogate models, applications in computational fluid dynamics, diffusion models for turbulent/climate forecasting, and learned samplers for rare-event sampling in computational chemistry. The course emphasizes mathematical foundations, engineering practice, benchmarking, and real-world deployment obstacles — teaching students when AI methods actually improve scientific computing, and when they don't.

Related event: CMU Launches New Course 10-749: AI for Scientific Computing with Open Lecture Notes(6 posts)→

Original post →

More from Research

Research channel →