Prerequisites:

Mandatory:  Linear Algebra. Basic Probability Theory.

Recommended:  Signal Processing. Machine Learning.

Signal Processing is the science that deals with extraction of information from signals of various kinds. This has two distinct aspects -- characterization and categorization. Traditionally, signal characterization has been performed with mathematically-driven transforms, while categorization and classification are achieved using statistical tools.

Machine learning aims to design algorithms that learn about the state of the world directly from data.

A increasingly popular trend has been to develop and apply machine learning techniques to both aspects of signal processing, often blurring the distinction between the two.

This course discusses the use of machine learning techniques to process signals. We cover a variety of topics, from data driven approaches for characterization of signals such as audio including speech, images and video, and machine learning methods for a variety of speech and image processing problems.

Students from the previous iteration of the course (fall2009) worked on a number of excellent projects, several of which were submitted to conferences.

The planned course schedule will appear here shortly.

There will be several guest lectures. These will be announced as dates are finalized.

Grading will be based on performance in course assignments and a final project.

The projects page: We had a terrific crop of projects this year

Class 1, 24 Aug 2010	Introduction. Representing Sounds and Images.	Slides, Handout
Class 2, 26 Aug 2010	Introduction to Linear Algebra	Slides, Handout
Class 3, 31 Aug 2010	Introduction to Linear Algebra II	Slides, Handout	Homework 1
Class 4, 2 Sep 2010	DSP refresher: data parameterization	Slides, Handout	Additional material
Class 5, 7 Sep 2010	Eigen decomposition. Eigen faces. Face detection I.	Slides, Handout
Class 6, 9 Sep 2010	Project Ideas.	Slides, Handout	Additional material
Class 7, 14 Sep 2010	Boosting, Face detection.	Slides, Handout	Additional material
Class 8, 16 Sep 2010	Hidden Markov Model Basics, guest lecture by Raffay Hamid	No slides (blackboard class!)
Class 9, 21 Sep 2010	Expectation Maximization	Slides, Handout
Class 10, 23 Sep 2010	Towards Energy-Aware Facilities Through Minimally Intrusive Approaches (guest lecture by Maio Berges)	Slides
Class 11, 28 Sep 2010	Intel Open House
Class 12, 30 Sep 2010	Component Analysis, Part 1 (guest lecture, Fernando de la Torre)	Slides
Class 13, 5 Oct 2010	Component Analysis, Part 2 (guest lecture, Fernando de la Torre)	Slides
Class 14, 7 Oct 2010	Expectation Maximization (2), Clustering	Slides, Handout	Additional material
Class 15, 12 Oct 2010	HMMs -- part 2	Slides, Handout	Additional material
Class 16, 14 Oct 2010	Weighted finite state transducers, John McDonough	Slides not available
Class 17, 19 Oct 2010	Iris recognition (Guest lecture, Marios Savvides)	Slides not available
Class 18, 21 Oct 2010	No class
Class 19, 26 Oct 2010	Guest lecture, Roger Dannenberg	Slides	Additional material
Class 20, 28 Oct 2010	Automatic speech recognition	Slides
Class 21, 2 Nov 2010	Latent variable models.	Slides
Class 22, 4 Nov 2010	Speech synthesis	Slides	Additional material
Class 23, 9 Nov 2010	No class		Additional material
Class 24, 11 Nov 2010	Sound modification	Slides
Class 25, 16 Nov 2010	Kalman filtering	Slides
Class 26, 18 Nov 2010	Compressive sensing. Guest lecture, Petros Boufounos	Slides
Class 27, 23 Nov 2010	Extended kalman filtering	Slides
Class 28, 30 Nov 2010	No class	Slides
Class 29, 2 Dec 2010	Project Presentations	Slides