1 Sparsity Control for Robust Principal Component Analysis Gonzalo Mateos and Georgios B. Giannakis ECE Department, University of Minnesota Acknowledgments:

Slides:



Advertisements
Similar presentations
1 Closed-Form MSE Performance of the Distributed LMS Algorithm Gonzalo Mateos, Ioannis Schizas and Georgios B. Giannakis ECE Department, University of.
Advertisements

Distributed Nuclear Norm Minimization for Matrix Completion
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
Pixel Recovery via Minimization in the Wavelet Domain Ivan W. Selesnick, Richard Van Slyke, and Onur G. Guleryuz *: Polytechnic University, Brooklyn, NY.
A KTEC Center of Excellence 1 Pattern Analysis using Convex Optimization: Part 2 of Chapter 7 Discussion Presenter: Brian Quanz.
Manifold Sparse Beamforming
CMPUT 466/551 Principal Source: CMU
Chapter 6 Feature-based alignment Advanced Computer Vision.
Shape From Light Field meets Robust PCA
Chapter 2: Lasso for linear models
Brian Baingana, Gonzalo Mateos and Georgios B. Giannakis Dynamic Structural Equation Models for Tracking Cascades over Social Networks Acknowledgments:
Bayesian Robust Principal Component Analysis Presenter: Raghu Ranganathan ECE / CMR Tennessee Technological University January 21, 2011 Reading Group (Xinghao.
Robust Network Compressive Sensing Lili Qiu UT Austin NSF Workshop Nov. 12, 2014.
1 Morteza Mardani, Gonzalo Mateos and Georgios Giannakis ECE Department, University of Minnesota Acknowledgment: AFOSR MURI grant no. FA
Computer Vision Spring ,-685 Instructor: S. Narasimhan Wean 5403 T-R 3:00pm – 4:20pm Lecture #20.
Image Super-Resolution Using Sparse Representation By: Michael Elad Single Image Super-Resolution Using Sparse Representation Michael Elad The Computer.
Coefficient Path Algorithms Karl Sjöstrand Informatics and Mathematical Modelling, DTU.
Principal Component Analysis
1-norm Support Vector Machines Good for Feature Selection  Solve the quadratic program for some : min s. t.,, denotes where or membership. Equivalent.
Image Denoising via Learned Dictionaries and Sparse Representations
Face Recognition Jeremy Wyatt.
Regression III: Robust regressions
Mehdi Ghayoumi MSB rm 132 Ofc hr: Thur, a Machine Learning.
Chapter 6 Feature-based alignment Advanced Computer Vision.
1 Sparsity Control for Robustness and Social Data Analysis Gonzalo Mateos ECE Department, University of Minnesota Acknowledgments: Profs. Georgios B. Giannakis,
Summarized by Soo-Jin Kim
Principle Component Analysis (PCA) Networks (§ 5.8) PCA: a statistical procedure –Reduce dimensionality of input vectors Too many features, some of them.
Technion - Israel Institute of Technology Department of Electrical Engineering Advanced Topics in Computer Vision Course Presentation By Stav Shapiro.
1 Unveiling Anomalies in Large-scale Networks via Sparsity and Low Rank Morteza Mardani, Gonzalo Mateos and Georgios Giannakis ECE Department, University.
1 Exact Recovery of Low-Rank Plus Compressed Sparse Matrices Morteza Mardani, Gonzalo Mateos and Georgios Giannakis ECE Department, University of Minnesota.
Cs: compressed sensing
Handling Outliers and Missing Data in Statistical Data Models Kaushik Mitra Date: 17/1/2011 ECSU Seminar, ISI.
EMIS 8381 – Spring Netflix and Your Next Movie Night Nonlinear Programming Ron Andrews EMIS 8381.
Machine Learning Seminar: Support Vector Regression Presented by: Heng Ji 10/08/03.
Independent Component Analysis Zhen Wei, Li Jin, Yuxue Jin Department of Statistics Stanford University An Introduction.
ECE 8443 – Pattern Recognition LECTURE 10: HETEROSCEDASTIC LINEAR DISCRIMINANT ANALYSIS AND INDEPENDENT COMPONENT ANALYSIS Objectives: Generalization of.
Virtual Vector Machine for Bayesian Online Classification Yuan (Alan) Qi CS & Statistics Purdue June, 2009 Joint work with T.P. Minka and R. Xiang.
Efficient computation of Robust Low-Rank Matrix Approximations in the Presence of Missing Data using the L 1 Norm Anders Eriksson and Anton van den Hengel.
Dimensionality Reduction Motivation I: Data Compression Machine Learning.
ECE 8443 – Pattern Recognition ECE 8423 – Adaptive Signal Processing Objectives: ML and Simple Regression Bias of the ML Estimate Variance of the ML Estimate.
Inference of Poisson Count Processes using Low-rank Tensor Data Juan Andrés Bazerque, Gonzalo Mateos, and Georgios B. Giannakis May 29, 2013 SPiNCOM, University.
CSE 185 Introduction to Computer Vision Face Recognition.
Biointelligence Laboratory, Seoul National University
High-dimensional Error Analysis of Regularized M-Estimators Ehsan AbbasiChristos ThrampoulidisBabak Hassibi Allerton Conference Wednesday September 30,
Multi-area Nonlinear State Estimation using Distributed Semidefinite Programming Hao Zhu October 15, 2012 Acknowledgements: Prof. G.
B. Baingana, E. Dall’Anese, G. Mateos and G. B. Giannakis Acknowledgments: NSF Grants , , , , ARO W911NF
CpSc 881: Machine Learning
Bundle Adjustment A Modern Synthesis Bill Triggs, Philip McLauchlan, Richard Hartley and Andrew Fitzgibbon Presentation by Marios Xanthidis 5 th of No.
Rank Minimization for Subspace Tracking from Incomplete Data
Robust Principal Components Analysis IT530 Lecture Notes.
MACHINE LEARNING 7. Dimensionality Reduction. Dimensionality of input Based on E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)
1 Robust Nonparametric Regression by Controlling Sparsity Gonzalo Mateos and Georgios B. Giannakis ECE Department, University of Minnesota Acknowledgments:
1 Consensus-Based Distributed Least-Mean Square Algorithm Using Wireless Ad Hoc Networks Gonzalo Mateos, Ioannis Schizas and Georgios B. Giannakis ECE.
Collaborative filtering applied to real- time bidding.
Jianchao Yang, John Wright, Thomas Huang, Yi Ma CVPR 2008 Image Super-Resolution as Sparse Representation of Raw Image Patches.
Dimension reduction (1) Overview PCA Factor Analysis Projection persuit ICA.
Martina Uray Heinz Mayer Joanneum Research Graz Institute of Digital Image Processing Horst Bischof Graz University of Technology Institute for Computer.
Introduction to several works and Some Ideas Songcan Chen
1 Dongheng Sun 04/26/2011 Learning with Matrix Factorizations By Nathan Srebro.
Efficient non-linear analysis of large data sets
Jinbo Bi Joint work with Jiangwen Sun, Jin Lu, and Tingyang Xu
Jeremy Watt and Aggelos Katsaggelos Northwestern University
ROBUST SUBSPACE LEARNING FOR VISION AND GRAPHICS
Outlier Processing via L1-Principal Subspaces
USPACOR: Universal Sparsity-Controlling Outlier Rejection
Blind Signal Separation using Principal Components Analysis
Sparse Regression-based Hyperspectral Unmixing
Sparse Principal Component Analysis
Introduction to Sensor Interpretation
Introduction to Sensor Interpretation
Presentation transcript:

1 Sparsity Control for Robust Principal Component Analysis Gonzalo Mateos and Georgios B. Giannakis ECE Department, University of Minnesota Acknowledgments: NSF grants no. CCF , EECS Asilomar Conference November 10, 2010

2 2 Principal Component Analysis Our goal: robustify PCA by controlling outlier sparsity Motivation: (statistical) learning from high-dimensional data Principal component analysis (PCA) [Pearson ’ 1901]  Extraction of low-dimensional data structure  Data compression and reconstruction  PCA is non-robust to outliers [Jolliffe ’ 86] DNA microarray Traffic surveillance

3 3 Our work in context Robust PCA  Robust covariance matrix estimators [Campbell ’ 80], [Huber ’ 81]  Computer vision [Xu-Yuille ’ 95], [De la Torre-Black ’ 03]  Low-rank matrix recovery from sparse errors [Wright et al ’ 09] Huber ’ s M-class and sparsity in linear regression [ Fuchs ’ 99] Contemporary applications  Anomaly detection in IP networks [Huang et al ’ 07], [Kim et al ’ 09]  Video surveillance, e.g., [Oliver et al ’ 99] OriginalRobust PCA `Outliers ’

4 4 PCA formulations Training data: Minimum reconstruction error: Dimensionality reduction operator Reconstruction operator Maximum variance: Factor analysis model: Solution:

5 5 Robustifying PCA Least-trimmed squares (LTS) regression [Rousseeuw ’ 87] (LTS PCA) LTS-based PCA for robustness is the -th order statistic among Trimming constant determines breakdown point Q: How should we go about minimizing ? (LTS PCA) is nonconvex; existence of minimizer(s)? A : Try all subsets of size, solve, and pick the best Simple but intractable beyond small problems

6 6 Modeling outliers Remarks  and are unknown  If outliers sporadic, then vector is sparse! Introduce auxiliary variables s.t. inlier outlier  Inliers obey ; outliers something else  Inlier noise: are zero-mean i.i.d. random vectors Natural (but intractable) estimator

7 7 LTS PCA as sparse regression Lagrangian form  Tuning controls sparsity in, thus number of outliers (P0) J ustifies the model and its estimator (P0); ties sparsity with robustness Proposition 1: If solves (P0) with chosen such that, then solves (LTS PCA) too.

8 Just relax! (P0) is NP-hard relax (P2) Q: Does (P2) yield robust estimates ? A: Yap! Huber estimator is a special case  Role of sparsity controlling is central

9 Entrywise outliers Use -norm regularization (P1) OriginalRobust PCA (P2)Robust PCA (P1) Outlier pixels Entire image rejected Outlier pixels rejected

10 Alternating minimization (P1)  update: reduced-rank Procrustes rotation  update: coordinatewise soft-thresholding Proposition 2: Alg. 1 ’ s iterates converge to a stationary point of (P1).

11 Refinements Nonconvex penalty terms approximate better in (P0) Options: SCAD [Fan-Li ’ 01], or sum-of-logs [Candes etal ’ 08] Iterative linearization-minimization of around Iteratively reweighted version of Alg. 1 Warm start: solution of (P1) or (P2) Bias reduction in (cf. weighted Lasso [Zou ’ 06]) Discard outliers identified in Re-estimate missing data problem

12 Online robust PCA Motivation: Real-time data and memory limitations Exponentially-weighted robust PCA Approximation [Yang ’ 95] At time, do not re-estimate past outlier vectors

13 Video surveillance OriginalPCARobust PCA `Outliers ’ Data:

14 Online PCA in action Angle between C(n) and C Inliers: Outliers: Figure of merit: angle between and

15 Concluding summary Sparsity control for robust PCA  LTS PCA as -(pseudo)norm regularized regression (NP-hard)  Relaxation (group)-Lassoed PCA M-type estimator  Sparsity controlling role of central Tests on real video surveillance data for anomaly extraction Batch and online robust PCA algorithms i) Outlier identification, ii) Robust subspace tracking Refinements via nonconvex penalty terms Ongoing research Preference measurement: conjoint analysis and collaborative filtering Robustifying kernel PCA and blind dictionary learning