Generative models are becoming powerful tools not only in AI, but also in science. In physics, they can help simulate and analyze complex systems that are difficult to study with traditional methods. This seminar introduces the intersection of machine learning, generative modeling, and physical systems, with a focus on applications in statistical physics and quantum field theory.
We will discuss how models such as autoregressive models, GANs, normalizing flows, and diffusion models can be used in scientific settings, and why physics provides a uniquely challenging testbed for modern ML. Topics include symmetry, physical constraints, scalability, and scientific interpretability. The course is designed for bachelor’s and master’s students in machine learning or computer science who are interested in research at the boundary of AI and science.
The seminar assumes some familiarity with generative models and basic machine learning. Prior physics knowledge is not required; the necessary background will be introduced along the way.
- Trainer/in: Ankur Singha