PHYS 598CPA, Computational Physics and Astrophysics, offered by Prof. Charles Gammie in Fall 2024

This fall a course PHYS 598CPA, Computational Physics and Astrophysics, will be offered by Prof. Charles Gammie.

In the future this course will be offered as PHYS 530, Computational Physics, and the syllabus this semester will reflect this change. The course description and brief explanation are provided below.

Course Description: Core computational techniques widely used in the physical sciences. Numerical differentiation, integration, solution of differential equations. Convergence, truncation error, and roundoff error. Numerical solution of hyperbolic, elliptic, and parabolic differential equations. Monte Carlo techniques. Elements of analysis, time series analysis, statistical models, and elementary applications of machine learning. Special topics.

Computational methods play a crucial role in modern physics research, on par with the analytic methods covered in the department's two-semester graduate sequence. Many graduate students encounter numerical techniques informally through their research groups, often using prepackaged software, without the chance to thoroughly understand the underlying algorithms or their limitations. This course offers an introduction to these computational methods, focusing on algorithms for solving physics-related problems. Topics include techniques for solving boundary value problems necessary for calculating scattering cross-sections, methods for solving coupled hyperbolic and elliptic PDEs that describe self-gravitating fluids, applications of the Metropolis algorithm to the Ising problem, and data analysis techniques such as Gaussian process regression and power spectrum estimation.