A Quick Practical Guide to PCA and ICA Ted Brookings, UCSB Physics 11/13/06.

Slides:



Advertisements
Similar presentations
FMRI Methods Lecture 10 – Using natural stimuli. Reductionism Reducing complex things into simpler components Explaining the whole as a sum of its parts.
Advertisements

Independent Component Analysis
Independent Component Analysis: The Fast ICA algorithm
Noise & Data Reduction. Paired Sample t Test Data Transformation - Overview From Covariance Matrix to PCA and Dimension Reduction Fourier Analysis - Spectrum.
Color Imaging Analysis of Spatio-chromatic Decorrelation for Colour Image Reconstruction Mark S. Drew and Steven Bergner
1er. Escuela Red ProTIC - Tandil, de Abril, 2006 Principal component analysis (PCA) is a technique that is useful for the compression and classification.
Independent Component Analysis & Blind Source Separation
Independent Component Analysis (ICA)
Dimensional reduction, PCA
Spike-triggering stimulus features stimulus X(t) multidimensional decision function spike output Y(t) x1x1 x2x2 x3x3 f1f1 f2f2 f3f3 Functional models of.
Independent Component Analysis & Blind Source Separation Ata Kaban The University of Birmingham.
Face Recognition Jeremy Wyatt.
Independent Component Analysis (ICA) and Factor Analysis (FA)
Bayesian belief networks 2. PCA and ICA
Some Statistics Stuff (A.K.A. Shamelessly Stolen Stuff)
ICA Alphan Altinok. Outline  PCA  ICA  Foundation  Ambiguities  Algorithms  Examples  Papers.
Laurent Itti: CS599 – Computational Architectures in Biological Vision, USC Lecture 7: Coding and Representation 1 Computational Architectures in.
Principal Component Analysis Principles and Application.
Dan Simon Cleveland State University
Principal Component Analysis. Philosophy of PCA Introduced by Pearson (1901) and Hotelling (1933) to describe the variation in a set of multivariate data.
Survey on ICA Technical Report, Aapo Hyvärinen, 1999.
Presented By Wanchen Lu 2/25/2013
Independent Components Analysis with the JADE algorithm
Principal Components Analysis BMTRY 726 3/27/14. Uses Goal: Explain the variability of a set of variables using a “small” set of linear combinations of.
ERP DATA ACQUISITION & PREPROCESSING EEG Acquisition: 256 scalp sites; vertex recording reference (Geodesic Sensor Net)..01 Hz to 100 Hz analogue filter;
Multivariate Approaches to Analyze fMRI Data Yuanxin Hu.
ECE 8443 – Pattern Recognition LECTURE 03: GAUSSIAN CLASSIFIERS Objectives: Normal Distributions Whitening Transformations Linear Discriminants Resources.
Independent Component Analysis on Images Instructor: Dr. Longin Jan Latecki Presented by: Bo Han.
Heart Sound Background Noise Removal Haim Appleboim Biomedical Seminar February 2007.
Additive Data Perturbation: data reconstruction attacks.
Independent Component Analysis Zhen Wei, Li Jin, Yuxue Jin Department of Statistics Stanford University An Introduction.
Principal Component Analysis Bamshad Mobasher DePaul University Bamshad Mobasher DePaul University.
Data Reduction. 1.Overview 2.The Curse of Dimensionality 3.Data Sampling 4.Binning and Reduction of Cardinality.
N– variate Gaussian. Some important characteristics: 1)The pdf of n jointly Gaussian R.V.’s is completely described by means, variances and covariances.
ECE 8443 – Pattern Recognition LECTURE 10: HETEROSCEDASTIC LINEAR DISCRIMINANT ANALYSIS AND INDEPENDENT COMPONENT ANALYSIS Objectives: Generalization of.
Descriptive Statistics vs. Factor Analysis Descriptive statistics will inform on the prevalence of a phenomenon, among a given population, captured by.
A note about gradient descent: Consider the function f(x)=(x-x 0 ) 2 Its derivative is: By gradient descent (If f(x) is more complex we usually cannot.
ECE 8443 – Pattern Recognition ECE 8423 – Adaptive Signal Processing Objectives: ML and Simple Regression Bias of the ML Estimate Variance of the ML Estimate.
Computational Intelligence: Methods and Applications Lecture 8 Projection Pursuit & Independent Component Analysis Włodzisław Duch Dept. of Informatics,
SEM Basics 2 Byrne Chapter 2 Kline pg 7-15, 50-51, ,
Lecture 3 BME452 Biomedical Signal Processing 2013 (copyright Ali Işın, 2013) 1 BME452 Biomedical Signal Processing Lecture 3  Signal conditioning.
PCA vs ICA vs LDA. How to represent images? Why representation methods are needed?? –Curse of dimensionality – width x height x channels –Noise reduction.
ECE 8443 – Pattern Recognition ECE 8527 – Introduction to Machine Learning and Pattern Recognition LECTURE 12: Advanced Discriminant Analysis Objectives:
Principal Component Analysis (PCA)
MACHINE LEARNING 7. Dimensionality Reduction. Dimensionality of input Based on E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)
Independent Component Analysis Independent Component Analysis.
Feature Selection and Extraction Michael J. Watts
Feature Extraction 主講人:虞台文. Content Principal Component Analysis (PCA) PCA Calculation — for Fewer-Sample Case Factor Analysis Fisher’s Linear Discriminant.
Introduction to Independent Component Analysis Math 285 project Fall 2015 Jingmei Lu Xixi Lu 12/10/2015.
An Introduction of Independent Component Analysis (ICA) Xiaoling Wang Jan. 28, 2003.
Analysis of FMRI Data: Principles and Practice Robert W Cox, PhD Scientific and Statistical Computing Core National Institute of Mental Health Bethesda,
Feature Extraction 主講人:虞台文.
Multivariate statistical methods. Multivariate methods multivariate dataset – group of n objects, m variables (as a rule n>m, if possible). confirmation.
Unsupervised Learning II Feature Extraction
Part 3: Estimation of Parameters. Estimation of Parameters Most of the time, we have random samples but not the densities given. If the parametric form.
Principal Component Analysis
LECTURE 11: Advanced Discriminant Analysis
Brain Electrophysiological Signal Processing: Preprocessing
Principal Component Analysis (PCA)
Dimension Reduction via PCA (Principal Component Analysis)
Application of Independent Component Analysis (ICA) to Beam Diagnosis
Bayesian belief networks 2. PCA and ICA
Techniques for studying correlation and covariance structure
Descriptive Statistics vs. Factor Analysis
X.1 Principal component analysis
Learning Theory Reza Shadmehr
Feature Selection Methods
Principal Component Analysis
Mathematical Foundations of BME
Computational Intelligence: Methods and Applications
Presentation transcript:

A Quick Practical Guide to PCA and ICA Ted Brookings, UCSB Physics 11/13/06

Blind Source Separation Suppose we have a data set that  Has many independent components or channels  Audio track recorded from multiple microphones  Series of brain images with multiple voxel  We believe is driven by several independent processes  Different people speaking into microphones  Different neuronal processes occurring within the brain  We have no a priori notion of what those processes look like Our goal is to figure out what the different processes are by grouping together data that is correlated

Our Simple Example  Driven by two sin signals with different frequencies  100 Sample Times  200 Channels:  150 are a linear combination of Signal1 and Signal2, with Poisson noise  50 are Pure Poisson noise

PCA (Principle Component Analysis)  Linear transform ---chooses a new basis  Perpendicular  First component explains the most variance, second component explains the most remaining variance, etc. Finds a weight matrix W, and set of signals S, that approximate the data X: X = W * S The weight matrix is the eigenvectors of the correlation matrix, so the eigenvalues provide the order of components Image from:

Spelling Things Out The meaning of the basis equation: e.g. if W =.6 and W =.2, then X =.6 S +.2 S. That is, X is actually being generated (at least partly) by the processes S and S X is typically a time series ---that is, X is measured at discrete intervals. However, our basis doesn’t change, because the fundamental processes that are at work is presumed to be constant. Because of this, W is constant in time, and S changes with time. The end result of PCA is then S(t), and W, which tells us the activity of each component, and how to generate the original data from the components.

PCA Results Unsurprisingly, PCA discovers two dominant components We might expect trouble here: PCA will probably go diagonal

PCA Results  Oops! The signals are mixed.  But… They’re a lot cleaner, because PCA has removed a lot of gaussian noise

ICA (Independent Component Analysis)  Linear transform ---chooses a new basis  NOT Perpendicular  The basis is chosen to be maximally-independent  There is no particular ordering of the basis vectors

Er… “Maximally Independent”? Correlated:Uncorrelated: Technical, and the definition depends somewhat on the algorithm being used. Ultimately boils down to cross-correlations. If two variables are uncorrelated, they are independent. Images from web page by Aapo Hyvärinen,

Requirements  At most one gaussian-distributed element of data  The number of independent data must be greater than the number of components: m > n. E.g. number of microphones greater than number of voices.

ICA Results Ick! Might have expected this, because there’s a ton of gaussian noise in the system.

Do ICA on the Results of PCA!  PCA cleans up the gaussian noise (and reduces the dimension).  Most PCA packages incorporate PCA or some other preprocessing for this reason.  ICA picks the basis that is maximally independent.

For More Info Check out Wikipedia (seriously). The articles on PCA/ICA  Are actually good.  Provide links to software packages for C++, Java, Matlab, etc. See especially FastICA.  Many of the external links provide good overviews as well.

The Aftermath… Great! Now that we have what we’ve always wanted (a list of “components”) what do we do with them? Since ICA is “blind” it doesn’t tells us much about the components. We may simply be interested in data reduction, or categorizing the mechanisms at work. We may be interested in components that correlate with some signal that we drove the experiment with.