Found 5 relevant results in 2.04s where lecturer="Yuansi Chen"
This is a master/PhD level course to prepare students for academic research in Markov chain Monte Carlo sampling algorithms, diffusion generative models and the associated theory. We cover basic sampling algorithms, provide geometric intuitions and demonstrate theoretical frameworks to analyze their computational and statistical performance.
This is a master/PhD level course to prepare students for academic research in Markov chain Monte Carlo sampling algorithms, diffusion generative models and the associated theory. We cover basic sampling algorithms, provide geometric intuitions and show frameworks for proving results in log-concave sampling.
Empirical process theory provides a rich toolbox for studying the properties of empirical risk minimizers, such as least squares and maximum likelihood estimators, support vector machines, etc.
Markov semigroups provide very general models and tools in the analysis of time evolution phenomena and dynamical systems.This course is a modern overview on probabilistic and geometric aspects of Markov semigroups and associated functional inequalities such as log-Sobolev inequalities.
Through reading a series of research articles on Bayesian computation, we first review several basic types of posterior sampling problems that arise in Bayesian statistical inference. Then we discuss how tailored MCMC sampling algorithms were designed to provide efficient sampling for each type of problem, the intuition behind them and theoretical justifications of their computational complexity.