Qing Feng Joint Work with JS Marron Jan Hannig Date 20140925 1 Era Challenge 2 Data Challenges MultiBlock data X Y S ubjects Feature Set 1 Feature Set 2 R apid growth of sources to obtain data ID: 788591
Download The PPT/PDF document "Non-iterative Joint and Individual Varia..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Non-iterative Joint and Individual Variation Explained
Qing Feng
Joint Work with J.S. Marron, Jan HannigDate: 2014/09/25
1
Slide2Era Challenge
2
Slide3Data Challenges
Multi-Block data
XY
S
ubjects
Feature Set 1
Feature Set 2
R
apid growth of sources to obtain data
High volume of available feature information
A variety of feature sets
3
Slide4Data Challenges
Multi-Block data challenges
High DimensionalityVariation Explanation
Heterogeneity
Singular value decomposition on concatenated data matrix
?
Singular value decomposition on Individual data blocks
?
Pre-transformation
?
4
Slide5Data Challenges
Toy Example
==++
+
+
Rank=1
Rank=1
X
Y
Standard Gaussian Random
M
atrix
5
Slide6Data Challenges
Simple concatenation
=
+
Low-rank SVD Approximation
Residual
Merged Data
6
Slide7Goal
Insightful Variation Decomposition
X =
Y
=
+
+
+
+
Joint Structure
Individual Structure
Residual
7
Slide8Essential Tool
Singular Value Decomposition (SVD) on
n
ull
Takeaway
Row space can be considered as ‘”Covariate” space
Singular values indicate “importance” of each covariate
8
Slide9Definition
Joint Structure
=
=
9
Slide10Definition
Individual structure
=
=
10
Slide11Definition
Residual
=
=
11
Slide12Outline of Implementation
Obtain covariate spaces of each structure
Recover structure matrices via projection
#
1 De-Noise
#
2
De-Noise
12
Slide13#1 De-noise
Extract signal based on singular values Perform for each block individually
Data Block X
Data Block Y
# 1 thresholds for each data block
13
Slide14#1 De-noise
#1 threshold selection[
Johnstone 2001] Limiting distribution of largest Eigen-value of covariance of standard Gaussian random matrix
In which,
follows Tracy-
Widom
law of distribution and
#1 threshold
14
Slide15#2 De-noise
Union of null spacesFrom noiseless case
Union
contains everything
but
row space
Null space of data X, Null(X) contains
Individual row space of Y,
Grey noise
row space
Null space of data Y, Null(Y) contains
Individual row space of X,
Grey noise
row space
15
Slide16#2 De-noise
Union of null spaces(UN) under noiseDirection spaces
Leaking joint signal becomes noiseEstimated signal direction deviates from direction in real row spaceSignal to ratio influences the angle
Joint in Y
Joint in X
Noise in X
Noise in
Y
Direction of leaking joint signal in UN
Direction of noise signal in UN
Paired in UN
16
Slide17#2 De-noise
Union of null spaces under noiseIdentify components via singular value plot
Detect pairs of basis in two estimated row spacesAngle distinguish joint from individual components
Noisy direction
#2 threshold
Paired basis
Calculate angle
Individual direction
17
Slide18#2 De-noise
#2 threshold selection
are estimation from [
Shablin
2010]
Suppose
as
,
In which,
18
Slide19#2 De-noise
Connection between #1 and #2 thresholds
What if
are unknown ?
Multi-scale
aspect comes!
19
Slide20Reconstruction
Project data blocks to each row spaces
Obtain estimations of each structure matricesTimes loading, singular value and basis in row together to recover each component
Data Block
Row space
Get Loading and singular value matrices
20
Slide21Toy Example
Data visualization
==++
+
+
Rank=1
Rank=1
X
Y
Standard Gaussian Random
M
atrix
21
Slide22Toy Example
#1 De-noise
Data Block XData Block Y
# 1 thresholds for each data block
22
Slide23Toy Example
#2 De-noise
Noisy direction
#2 threshold
Individual direction
Joint direction
23
Slide24Toy Example
Reconstruction
=++=
+
+
Joint
Individual
Residual
X
Y
24
Slide25Spanish Mortality
Male block Versus Female block
Age as features
Age as features
Years as Subjects
Spanish Male
Spanish Female
+
-
25
Slide26Spanish Male
26
Slide27Spanish Female
27
Slide28Spanish Mortality
#1 De-noise
Data Block MaleRank=7Data Block Female
Rank=6
(Log of singular value)
Eyeball # 1 thresholds
28
Slide29Spanish Mortality
#2 De-noise
#2 thresholdRank of Joint structure=4
Joint directions
Noisy direction
Individual directions
29
Slide30Joint Variation - Male
30
Slide31Joint Variation - Female
31
Slide32Individual Variation - Male
Spanish Civil War
32
Slide33Individual Variation - Female
Flue epidemic
33
Slide34Thank you!
34