Model-Based Estimation and Signal Analysis

Spring Semester 2024

Prof. Hans-Andrea Loeliger

Location: ML F 39
Lectures: Friday, 14:15 – 16:00
Tutorial sessions: 16:15 – 18:00
Assistants: Tianyang Wang and Yunpeng Li

Lecture Notes, Problems, and Solutions (login)

Description

The course develops a selection of topics pivoting around state space models, factor graphs, and pertinent algorithms for estimation, model fitting, and learning:

• hidden-Markov models
• factor graphs and message passing algorithms
• linear state space models, Kalman filtering and smoothing, recursive least squares
  
• Gibbs sampling, particle filter
• recursive local model fitting for signal analysis
  
• parameter learning by expectation maximization
• iterative algorithms for model fitting with Lp costs, sparsity, and discrete variables

Prerequisites

Solid mathematical foundations (especially in probability, estimation, and linear algebra), as provided by the course "Introduction to Estimation and Machine Learning".

Lecture Notes

Complete lecture notes (in English) will be handed out as the course progresses.

Examination

Written (in English).