Iterative Stochastic Numerical Methods for Statistical Sampling

Explore the foundations of molecular simulation and stochastic algorithms for Bayesian inference in this lecture by Professor Ben Leimkuhler at the Alan Turing Institute. Delve into the design, analysis, and implementation of algorithms for time-dependent phenomena and modeling in engineering and sciences. Discover how molecular dynamics temperature controls can be applied to stabilize sampling-based parameterization schemes in data analytics. Examine the disruption of traditional mathematical models by data science, incorporating massive data sets and empirical laws. Learn about the integration of Bayesian inference methods with physical law-based models, dynamical principles, and geometric constraints. Gain insights into the interplay between naive data science approaches and models informed by physical laws and mathematical structures. This 59-minute talk bridges the gap between molecular science and data analytics, offering valuable perspectives on the evolving landscape of computational methods in scientific research.

Iterative stochastic numerical methods for statistical sampling: Professor Ben Leimkuhler