Course Materials

We will broadly categorize the key procedures we apply to signals as Learning representations, Modelling, Classification and Prediction. In the list below we have indicated the categories that apply to the topics of each lectures. Some of the lectures relate to basics, and have been marked as such.

Date Lecture topics Additional Materials
1 Sep 2016 Intro: Introduction to course, notion of a signal, basic digital representation of data, e.g. speech, images, other types of data [slides]  
6 Sep 2016 Basics: Fundamentals of Linear Algebra, part 1: Spaces, algebraic operations and their interpretations, projections [slides]  
8 Sep 2016 Basics: Fundamentals of Linear Algebra, part 2: Types of operations, Eigen decompositions, SVD, Vector/Matrix calculus [slides]  
13 Sep 2016 Basics: Basics of function optimization [slides]  
15 Sep 2016 Discussion of project ideas [slides]  
20 Sep 2016 Representation: Deterministic representations of signals [slides]  
22 Sep 2016 Representation: Data-driven representations: Eigen representations, PCA, Properties of these representations [slides]  
27 Sep 2016 Classification: An introduction to classification: binary classification, Boosting, and an application to face detection in images [slides]  
29 Sep 2016 Representation: Data-driven representations : NMF, types of NMF, overcomplete representations, sparsity, applications: signal representation, signal separation [slides]  
4 Oct 2016 Guest Lecture (Roger Dannenberg): Music Understanding [slides]  
6 Oct 2016 Representation: Data-driven representations: Independent Component Analysis, ICA for signal representations and denoising, example applications [slides]  
11 Oct 2016 Guest lecture (Felix Xu and Marios Savvides): Correlation Filters[slides]  
13 Oct 2016 Representation/modelling: Clustering[slides]  
18 Oct 2016 Representation: Sparse representations and overcomplete dictionaries[slides]  
20 Oct 2016 Representation: Compressive sensing[slides]  
25 Oct 2016 Modelling: Regression[slides]  
27 Oct 2016 Modelling: Expectation Maximization[slides]  
1 Nov 2016 Guest Lecture (Taylor Berg-Kirkpatrick): Music Transcription [slides]  
3 Nov 2016 Representation/Modelling: Factor Analysis (Najim)[slides]  
8 Nov 2016 Representation/Modelling: Supervised representations: CCA and LDA [slides]  
10 Nov 2016 Modelling: Hidden Markov Models[slides]  
15 Nov 2016 Modelling/Prediction: Prediction, Tracking, Kalman Filtering [slides]  
29 Nov 2016 Modelling/Prediction: Neural Networks Part I [slides]  
1 Dec 2016 Modelling/Prediction: Neural Networks Part II [slides]