Delve into the foundations of probabilistic state space modeling in this comprehensive talk presented by Scott Linderman, Assistant Professor in the Statistics Department at Stanford University. Explore essential tools for analyzing sequential data, including neural and behavioral time series, and gain insights into the latent states and dynamics underlying high-dimensional measurements. Learn about discrete and continuous state space models such as Hidden Markov Models and linear Gaussian dynamical systems, as well as more complex models like switching linear and nonlinear dynamical systems. Discover both exact and approximate algorithms for learning (parameter estimation) and inference (state estimation). Benefit from a blend of mathematical derivations and practical code demonstrations using the new dynamax library. This talk is suitable for those with a basic understanding of linear algebra, multivariate calculus, and fundamental probability concepts.
Overview
Syllabus
Talk 1: Nuts and Bolts of Modern State Space Models - Part I
Taught by
Georgia Tech Research
Related Courses
-
State space Approach to Control System Analysis and Design
-
Introduction to Scientific Machine Learning
-
Time Series Modelling and State Space Models - Professor Chris Williams, University of Edinburgh
-
Time Series Class - Part 2 - Professor Chris Williams, University of Edinburgh
-
An Introduction to State-Space Models, Particle Filters, and Sequential Monte Carlo - Lecture 1
-
Time Series Class - Part 1 - Dr Ioannis Papastathopoulos, University of Edinburgh
