Rank Minimization for Subspace Tracking from Incomplete Data

Slides:



Advertisements
Similar presentations
BiG-Align: Fast Bipartite Graph Alignment
Advertisements

1 Closed-Form MSE Performance of the Distributed LMS Algorithm Gonzalo Mateos, Ioannis Schizas and Georgios B. Giannakis ECE Department, University of.
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.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Solving POMDPs Using Quadratically Constrained Linear Programs Christopher Amato.
T HE POWER OF C ONVEX R ELAXATION : N EAR - OPTIMAL MATRIX COMPLETION E MMANUEL J. C ANDES AND T ERENCE T AO M ARCH, 2009 Presenter: Shujie Hou February,
Manifold Sparse Beamforming
Wangmeng Zuo, Deyu Meng, Lei Zhang, Xiangchu Feng, David Zhang
Javad Lavaei Department of Electrical Engineering Columbia University Graph-Theoretic Algorithm for Nonlinear Power Optimization Problems.
Shape From Light Field meets Robust PCA
Probabilistic Clustering-Projection Model for Discrete Data
More MR Fingerprinting
Ilias Theodorakopoulos PhD Candidate
ECE 8443 – Pattern Recognition ECE 8423 – Adaptive Signal Processing Objectives: Newton’s Method Application to LMS Recursive Least Squares Exponentially-Weighted.
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
Modern Sampling Methods Summary of Subspace Priors Spring, 2009.
Image Denoising via Learned Dictionaries and Sparse Representations
© 2011 IBM Corporation IBM Research SIAM-DM 2011, Mesa AZ, USA, Non-Negative Residual Matrix Factorization w/ Application to Graph Anomaly Detection Hanghang.
Computing Sketches of Matrices Efficiently & (Privacy Preserving) Data Mining Petros Drineas Rensselaer Polytechnic Institute (joint.
Mehdi Ghayoumi MSB rm 132 Ofc hr: Thur, a Machine Learning.
1 Dot Plots For Time Series Analysis Dragomir Yankov, Eamonn Keogh, Stefano Lonardi Dept. of Computer Science & Eng. University of California Riverside.
Sparsity-Aware Adaptive Algorithms Based on Alternating Optimization and Shrinkage Rodrigo C. de Lamare* + and Raimundo Sampaio-Neto * + Communications.
Online Dictionary Learning for Sparse Coding International Conference on Machine Learning, 2009 Julien Mairal, Francis Bach, Jean Ponce and Guillermo Sapiro.
1 Sparsity Control for Robustness and Social Data Analysis Gonzalo Mateos ECE Department, University of Minnesota Acknowledgments: Profs. Georgios B. Giannakis,
Brian Baingana, Gonzalo Mateos and Georgios B. Giannakis A Proximal Gradient Algorithm for Tracking Cascades over Networks Acknowledgments: NSF ECCS Grant.
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.
Matrix Completion IT530 Lecture Notes.
1 Exact Recovery of Low-Rank Plus Compressed Sparse Matrices Morteza Mardani, Gonzalo Mateos and Georgios Giannakis ECE Department, University of Minnesota.
1 Jorge Nocedal Northwestern University With S. Hansen, R. Byrd and Y. Singer IPAM, UCLA, Feb 2014 A Stochastic Quasi-Newton Method for Large-Scale Learning.
Recovering low rank and sparse matrices from compressive measurements Aswin C Sankaranarayanan Rice University Richard G. Baraniuk Andrew E. Waters.
EE369C Final Project: Accelerated Flip Angle Sequences Jan 9, 2012 Jason Su.
1 Sparsity Control for Robust Principal Component Analysis Gonzalo Mateos and Georgios B. Giannakis ECE Department, University of Minnesota Acknowledgments:
Online Learning for Collaborative Filtering
Scalable Symbolic Model Order Reduction Yiyu Shi*, Lei He* and C. J. Richard Shi + *Electrical Engineering Department, UCLA + Electrical Engineering Department,
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.
Direct Robust Matrix Factorization Liang Xiong, Xi Chen, Jeff Schneider Presented by xxx School of Computer Science Carnegie Mellon University.
Javad Lavaei Department of Electrical Engineering Columbia University Joint work with Somayeh Sojoudi and Ramtin Madani An Efficient Computational Method.
Inference of Poisson Count Processes using Low-rank Tensor Data Juan Andrés Bazerque, Gonzalo Mateos, and Georgios B. Giannakis May 29, 2013 SPiNCOM, University.
Javad Lavaei Department of Electrical Engineering Columbia University Convex Relaxation for Polynomial Optimization: Application to Power Systems and Decentralized.
Powerpoint Templates Page 1 Powerpoint Templates Scalable Text Classification with Sparse Generative Modeling Antti PuurulaWaikato University.
E XACT MATRIX C OMPLETION VIA CONVEX OPTIMIZATION E MMANUEL J. C ANDES AND B ENJAMIN R ECHT M AY 2008 Presenter: Shujie Hou January, 28 th,2011 Department.
Multi-area Nonlinear State Estimation using Distributed Semidefinite Programming Hao Zhu October 15, 2012 Acknowledgements: Prof. G.
Designing Games for Distributed Optimization Na Li and Jason R. Marden IEEE Journal of Selected Topics in Signal Processing, Vol. 7, No. 2, pp ,
Bundle Adjustment A Modern Synthesis Bill Triggs, Philip McLauchlan, Richard Hartley and Andrew Fitzgibbon Presentation by Marios Xanthidis 5 th of No.
Ariadna Quattoni Xavier Carreras An Efficient Projection for l 1,∞ Regularization Michael Collins Trevor Darrell MIT CSAIL.
Large-Scale Matrix Factorization with Missing Data under Additional Constraints Kaushik Mitra University of Maryland, College Park, MD Sameer Sheoreyy.
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.
Unsupervised Streaming Feature Selection in Social Media
Facets: Fast Comprehensive Mining of Coevolving High-order Time Series Hanghang TongPing JiYongjie CaiWei FanQing He Joint Work by Presenter:Wei Fan.
State-Space Recursive Least Squares with Adaptive Memory College of Electrical & Mechanical Engineering National University of Sciences & Technology (NUST)
Introduction to several works and Some Ideas Songcan Chen
Super-resolution MRI Using Finite Rate of Innovation Curves Greg Ongie*, Mathews Jacob Computational Biomedical Imaging Group (CBIG) University of Iowa.
Jinbo Bi Joint work with Tingyang Xu, Chi-Ming Chen, Jason Johannesen
Jinbo Bi Joint work with Jiangwen Sun, Jin Lu, and Tingyang Xu
Jeremy Watt and Aggelos Katsaggelos Northwestern University
Jeremy Watt and Aggelos Katsaggelos Northwestern University
ROBUST SUBSPACE LEARNING FOR VISION AND GRAPHICS
USPACOR: Universal Sparsity-Controlling Outlier Rejection
Sudocodes Fast measurement and reconstruction of sparse signals
High-Dimensional Matched Subspace Detection When Data are Missing
Optimal sparse representations in general overcomplete bases
Recursively Adapted Radial Basis Function Networks and its Relationship to Resource Allocating Networks and Online Kernel Learning Weifeng Liu, Puskal.
Outline Sparse Reconstruction RIP Condition
Sebastian Semper1 and Florian Roemer1,2
Presentation transcript:

Rank Minimization for Subspace Tracking from Incomplete Data Morteza Mardani, Gonzalo Mateos and Georgios Giannakis ECE Department, University of Minnesota Acknowledgment: AFOSR MURI grant no. FA9550-10-1-0567 Vancouver, Canada May 18, 2013 1 1

Learning from “Big Data” `Data are widely available, what is scarce is the ability to extract wisdom from them’ Hal Varian, Google’s chief economist Fast BIG Ubiquitous Revealing Productive Smart Messy 2 K. Cukier, ``Harnessing the data deluge,'' Nov. 2011. 2

Streaming data model Incomplete observations Sampling operator: Preference modeling Incomplete observations ? ? ? ? ? Sampling operator: lives in a slowly-varying low-dimensional subspace Goal: Given and estimate and recursively

Prior art (Robust) subspace tracking Batch rank minimization Projection approximation (PAST) [Yang’95] Missing data: GROUSE [Balzano et al’10], PETRELS [Chi et al’12] Outliers: [Mateos-Giannakis’10], GRASTA [He et al’11] Batch rank minimization Nuclear norm regularization [Fazel’02] Exact and stable recovery guarantees [Candes-Recht’09] Novelty: Online rank minimization Scalable and provably convergent iterations Attain batch nuclear-norm performance subspace tracking which has a rich litertature, PAST (projection approximation subspace tracking): GROUSE (Grassmanian rand-one update subspace estimation): inceremental gradient over the Grassmanian PETRELS: simple extension of PAST to account for missing data, knows the true rank, second order algorithm, original algorithms lack regularization and thus it is numerically unstable especially for large amounts of missing observations

Low-rank matrix completion Consider matrix , set Sampling operator Given incomplete (noisy) data (as) has low rank Goal: denoise observed entries, impute missing ones Nuclear-norm minimization [Fazel’02],[Candes-Recht’09] 5 5

Problem statement Available data at time t ? ? ? ? ? ? ? Goal: Given historical data , estimate from (P1) Challenge: Nuclear norm is not separable Variable count Pt growing over time Costly SVD computation per iteration 6 6

Separable regularization Key result [Burer-Monteiro’03] Pxρ ≥rank[X] New formulation equivalent to (P1) (P2) Nonconvex; reduces complexity: Proposition 1. If stationary pt. of (P2) and , then is a global optimum of (P1). 7 7

Online estimator Regularized exponentially-weighted LS estimator (0 < β ≤ 1 ) (P3) := Ct(L,Q) Alternating minimization (at time t) Step1: Projection coefficient updates Step2: Subspace update := gt(L[t-1],q) 8 8

Online iterations Attractive features ρxρ inversions per time, no SVD, O(Pρ3) operations (ind. of time) β=1: recursive least-squares; O(Pρ2) operations 9 9

Convergence As1) Invariant subspace and As2) Infinite memory β = 1 Proposition 2: If and are i.i.d., and c1) is uniformly bounded; c2) is in a compact set; and c3) is strongly convex w.r.t. hold, then almost surely (a. s.) Q1: is it reasonable to assume {y_t} is i.i.d. over time? ???? Q2: what is the general idea of the proof technique? asymptotically converges to a stationary point of batch (P2)

Optimality Q: Given the learned subspace and the corresponding is an optimal solution of (P1)? Proposition 3: If there exists a subsequence s.t. then satisfies the optimality conditions for (P1) as a. s. c1) a. s. c2) 11 11

Numerical tests Data , Optimality (β=1) Performance comparison (β=0.99, λ=0.1) (P1) Efficient for large-scale matrix completion Complexity comparison Algorithm 1 O(Pρ3) PETRELS O(Pρ2) GROUSE O(Pρ) 12 12

Tracking Internet2 traffic Goal: Given a small subset of OD-flow traffic-levels estimate the rest Traffic is spatiotemporally correlated Real network data Dec. 8-28, 2008; N=11, L=41, F=121, T=504 k=ρ=10, β=0.95 π=0.25 13 13 Data: http://www.cs.bu.edu/~crovella/links.html 13 13

Dynamic anomalography Estimate a map of anomalies in real time Streaming data model: Goal: Given estimate online when is in a low-dimensional space and is sparse ---- estimated ---- real M. Mardani, G. Mateos, and G. B. Giannakis, "Dynamic anomalography: Tracking network anomalies via sparsity and low rank," IEEE Journal of Selected Topics in Signal Process., vol. 7, pp. 50-66, Feb. 2013.

Conclusions Thank You! Track low-dimensional subspaces from Incomplete (noisy) high-dimensional datasets Online rank minimization Scalable and provably convergent iterations attaining batch nuclear-norm performance Viable alternative for large-scale matrix completion Extensions to the general setting of dynamic anomalography Future research Accelerated stochastic gradient for subspace update Adaptive subspace clustering of Big Data Thank You! 15 15