EE 378: Inference, Estimation and Information Processing

Announcements

• The graded final exam can be picked up from Kelly Yilmaz at Packard 259. [solutions]

• All homework solutions are posted.

• The scribe of lecture 17 is posted.

• Final exam will be on Wednesday, June 9, 8:30-11:30 AM, at 260-113.

• Homework 5 is posted, due on Friday, June 4, 5PM.

• There will be a review session on Friday, May 21, 11:00–12:00, at Hewlett 103.

• The TA evaluation form will be available until Monday, May 3.

• The course reader from EE378, Spring Quarter 2008-09 is posted.

• Email the TA with your choice of lecture scribing days.

• Please register on the CCNet website where we will post grades.

• The first lecture will be on Monday, March 29, 11:00–12:15, at Y2E2 111.

• Lectures will be made available in video via ClassX.

Course Overview

This is an experimental new format for EE378, formerly and formally known as statistical signal processing. While maintaining much of the core from the original version, we extend it to be an introduction to principles of inference, modeling and information processing, with some emphasis on information theoretic perspectives. Students will be exposed to basic principles, as well as their manifestation in the construction of – and performance analysis for – both contemporary and classical information processing algorithms. Themes include:

• Bayesian inference, estimation, and hypothesis testing

• Linear vs. non-linear, causal vs. non-causal, exact vs. approximate inference

• Hidden Markov processes and some more general graphical models

• Log-loss inference and MDL

• Information and estimation

• Prediction, filtering, and denoising (classical and modern)