This document summarizes David Gleich's presentation on skew-symmetric matrix completion for rank aggregation and other matrix computations. Gleich discussed how (1) many machine learning and data analysis tasks involve matrix computations, (2) rankings can be represented as skew-symmetric matrices, and (3) the problem of rank aggregation involves completing a partially observed skew-symmetric matrix. Gleich proposed solving this problem using nuclear norm regularization, which encourages low-rank solutions and can be solved efficiently using singular value thresholding. The nuclear norm formulation relaxes the hard matrix completion problem into a convex optimization problem that can recover rankings from partial information.
Skew-symmetric matrix completion for rank aggregation
1. Skew-symmetric matrix
completion for rank
aggregation !
and other matrix computations
DAVID F. GLEICH
PURDUE UNIVERSITY
COMPUTER SCIENCE DEPARTMENT
1/40
February 24 th , 12pm
Purdue ML Seminar
David Gleich, Purdue
2. Skew-symmetric matrix
completion for rank
aggregation !
and other matrix computations
DAVID F. GLEICH
PURDUE UNIVERSITY
COMPUTER SCIENCE DEPARTMENT
2/40
February 24 th , 12pm
Purdue ML Seminar
David Gleich, Purdue
3. Skew-symmetric matrix
completion for rank
aggregation !
and other matrix computations
DAVID F. GLEICH
PURDUE UNIVERSITY
COMPUTER SCIENCE DEPARTMENT
3/40
January 24 th , 12pm
Purdue ML Seminar
David Gleich, Purdue
9. A new matrix-based sensitivity
analysis of Google’s PageRank.
PageRank (I ↵P)x = (1 ↵)v
SimRank
Presented at" RAPr on Wikipedia
DiffusionRank
WAW2007, WWW2010
E [x(A)] Std [x(A)]
BlockRank
Published in the
United States IsoRank
United States
C:Living people C:Living people
TrustRank
J. Internet Mathematics
France ItemRank
C:Main topic classif.
ObjectRank
ProteinRank
Led to new results on United Kingdom C:Contents
uncertainty quantification in Germany C:Ctgs. by country
HostRank
physical simulations
published in SIAM J. Matrix
England
Canada
SocialPageRank
United Kingdom
France
Random walk with
Analysis and SIAM J.
Scientific Computing.
Japan
Poland
FoodRank
C:Fundamental
England
restart
Patent Pending
Australia
FutureRank
C:Ctgs. by topic
GeneRank
TwitterRank
Improved web-spam detection!
Gleich (Stanford) Random sensitivity Ph.D. Defense 23 / 41
Collaborators Paul Constantine, Gianluca Iaccarino (physical simulation)
10. j
Square s
2 F.L (Purdue)
vid Gleich
r
Network alignment INFORMS Semina
= (t, )
twork alignment
= t t
t
mm 40 60 80 100
A L B
NETWORK ALIGNMENT
X m ximize wT x + 2 xT Sx
T x + 1 xT Sx
40 j subject to Axw e, 2 {0, 1}
m ximize
x 2
S $ subject to Ax e
ng
ry Network alignment
2 {0, 1} problems
Sparse
10/40
Bayati, Gerritsen, Gleich, Saberi, and Wang, ICDM2009
UADRATIC ASSIGNMENT Bayati, Gleich, Saberi and Wang,often ignore
Sparse L Submitted
60
Southeast Ranking few exceptions).
Purdue ML Seminar
David Gleich, Purdue
Network alignment Workshop 11 / 29
11. Overlapping clusters!
for distributed computation
Andersen, Gleich, and Mirrokni, WSDM2012
2
Swapping Probability (usroads)
PageRank Communication (usroads)
Swapping Probability (web−Google)
1.5
PageRank Communication (web−Google)
Relative Work
1 Metis Partitioner
0.5
0
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Volume Ratio
11/40
How much more of the graph we need to store.
Purdue ML Seminar
David Gleich, Purdue
12. Local methods for massive Twee
network analysis
RESULTS – SLIDE THRE Gleich et al. "
MAIN J. Internet Mathematics, to appear.
TOP-K ALGORITHM FOR KATZ
Approximate
where is sparse
Keep sparse too
Ideally, don’t “touch” all of
David F. Gleich (Purdue) Univ. Chicago SSCS Seminar 34 of 47
Can solve these problemsGleich milliseconds even withICME la/opt seminar
David F. in (Sandia) 100M edges!
12/40
Purdue ML Seminar
David Gleich, Purdue
13. DAVID F. GLEICH (PURDUE) &
LEK-HENG LIM (UNIV. CHICAGO)
Rank
aggregation
13
Purdue ML Seminar
David Gleich, Purdue
14. Which is a better list of good DVDs?
Lord of the Rings 3: The Return of …
Lord of the Rings 3: The Return of …
Lord of the Rings 1: The Fellowship
Lord of the Rings 1: The Fellowship
Lord of the Rings 2: The Two Towers
Lord of the Rings 2: The Two Towers
Lost: Season 1
Star Wars V: Empire Strikes Back
Battlestar Galactica: Season 1
Raiders of the Lost Ark
Fullmetal Alchemist
Star Wars IV: A New Hope
Trailer Park Boys: Season 4
Shawshank Redemption
Trailer Park Boys: Season 3
Star Wars VI: Return of the Jedi
Tenchi Muyo!
Lord of the Rings 3: Bonus DVD
Shawshank Redemption
The Godfather
Standard " Nuclear Norm "
rank aggregation" based rank aggregation
(the mean rating)
(not matrix completion on the
14/40
netflix rating matrix)
Purdue ML Seminar
David Gleich, Purdue
15. Rank Aggregation
Given partial orders on subsets of items, rank aggregation
is the problem of finding an overall ordering.
Voting Find the winning candidate
Program committees Find the best papers given reviews
Dining Find the best restaurant in Chicago
15/40
Purdue ML Seminar
David Gleich, Purdue
16. Ranking is really hard
John Kemeny
Dwork, Kumar, Naor, !
Ken Arrow
Sivikumar
All rank aggregations
involve some measure of A good ranking is the
compromise
“average” ranking under a NP hard to compute
Kemeny’s ranking
16/40
permutation distance
Purdue ML Seminar
David Gleich, Purdue
17. Embody chair!
John Cantrell (flickr)
Given a hard problem,
what do you do?!
!
Numerically relax!!
!
It’ll probably be easier.
17/40
Purdue ML Seminar
David Gleich, Purdue
18. Suppose we had scores
Suppose we had scores
Let be the score of the ith movie/song/paper/team to rank
Suppose we can compare the ith to jth:
Then is skew-symmetric, rank 2.
Also works for with an extra log.
Numerical ranking is intimately intertwined
with skew-symmetric matrices
18/40
Kemeny and Snell, Mathematical Models in Social Sciences (1978)
David F. Gleich (Purdue) Purdue
KDD 2011 ML Seminar
David Gleich, Purdue
6/20
19. Using ratings as comparisons
Arithmetic Mean
Ratings induce
various skew-
symmetric matrices.
Log-odds
19/40
From David 1988 – The
Method of Paired Comparisons
Purdue ML Seminar
David Gleich, Purdue
20. Extracting the scores
Extracting the scores
Given with all entries, then 107
is the Borda
Movie Pairs
105
count, the least-squares
solution to
How many do we have? 101
Most.
101 105
Number of Comparisons
Do we trust all ?
Not really. Netflix data 17k movies,
500k users, 100M ratings–
99.17% filled
20/40
David F. Gleich (Purdue) Purdue
KDD 2011 ML Seminar
David Gleich, Purdue
8/20
21. Onlypartial info? COMPLETE IT!
Only partial info? Complete it!
Let be known for We trust these scores.
Goal Find the simplest skew-symmetric matrix that matches
the data
noiseless
noisy
21/40
Both of these are NP-hard too.
Purdue ML Seminar
David Gleich, Purdue
David F. Gleich (Purdue) KDD 2011 9/20
22. Solution GO NUCLEAR!
22/40
From a French nuclear test in 1970, imagePurdue ML Seminar
David Gleich, Purdue
from http://picdit.wordpress.com/2008/07/21/8-
insane-nuclear-explosions/
23. The nuclear norm
The nuclear norm!
The analog the 1-norm or -norm for matrices
The analog of of the 1-norm or ℓ������1for matrices.
For vectors For matrices
Let be the SVD.
is NP-hard while
is convex and gives the same best convex under-
answer “under appropriate estimator of rank on unit ball.
circumstances”
23/40
Purdue ML Seminar
David Gleich, Purdue
24. Only partial info? COMPLETE IT!
Only partial info? Complete it!
Let be known for We trust these scores.
Goal Find the simplest skew-symmetric matrix that matches
the data
NP hard
Heuristic
Convex
24/40
Purdue ML Seminar
David Gleich, Purdue
25. Solving the !
Solving theproblem
nuclear norm nuclear norm problem
Use a LASSO formulation 1.
2. REPEAT
3. = rank-k SVD of
4.
5.
6. UNTIL
Jain et al. propose SVP for
this problem without
25/40
Purdue ML Seminar
David Gleich, Purdue
26. Skew-symmetric SVDs
Skew-symmetric SVD
Let be an skew-symmetric matrix with
eigenvalues ,
where and . Then the SVD of is
given by
for and given in the proof.
Proof Use the Murnaghan-Wintner form and the SVD of a
2x2 skew-symmetric block
This means that SVP will give us the skew-
26/40
symmetric constraint “for free”
David F. Gleich (Purdue) KDD 2011 14/20
Purdue ML Seminar
David Gleich, Purdue
27. Only partial info? Complete it!
Let be known for We trust these score
Matrix completion
Goal Find the simplest skew-symmetric matrix that
the data
A fundamental
question is matrix
NP hard
completion is
when do these
problems have the
Convex
same solution?
27/40
David F. Gleich (Purdue) KDD 2011
Purdue ML Seminar
David Gleich, Purdue
28. indices. Instead we view the following theorem as providing
Fraction of trials recovered
1
intuition for the noisy problem.
0.8
Exact recovery results
Consider the operator basis for Hermitian matrices:
Exact recovery results
H = S [ K [ D where 0.4
0.6
p
S = {1/ 2(ei eT + ej eT ) : 1 i < j n};
David Gross showed how to recover Hermitian matrices. 0.2
p
j i
K = {ı/ 2(ei eT ej eT ) : 1we get n}; exact
i.e. the conditions under which i < j the
j i 0
2
10
T
D = {ei ei : 1 i n}. Gross, arXiv, 2010
Note that is Hermitian. Thus our new result!
Figure
T
Theorem 5. Let s be centered, i.e., s e = 0. Let Y = ity of
seT esT where ✓ = maxi s2 /(sT s) and ⇢ = ((maxi si )
i about
(mini si ))/ksk. Also, let ⌦ ⇢ H be a random set of elements both th
with size |⌦| O(2n⌫(1 + )(log n)2 ) where ⌫ = max((n✓ + §6.1 fo
1)/4, n⇢2 ). Then the solution of
6.1 R
minimize kXk⇤
The fi
⇤ ⇤
subject to trace(X W i ) = trace((ıY ) W i ), W i 2 ⌦ ability o
the nois
28/40
is equal to ıY with probability at least 1 n . with un
These a
The proof of this theorem follows directly by Theorem 4 if
Purdue ML Seminar
David Gleich, Purdue
Y = se
29. Recovery Discussion and Experiments
Confession If , then just look at differences from
a connected set. Constants? Not very good.
Intuition for the truth.
29/40
Purdue ML Seminar
David Gleich, Purdue
30. Recovery Discussion and Experiments
Recovery
Confession If Experiments
just look at differences from
, then
a connected set. Constants? Not very good.
Intuition for the truth.
30/40
David F. Gleich (Purdue) KDD 2011 16/20
Purdue ML Seminar
David Gleich, Purdue
31. The ranking algorithm
Algorithm
The Ranking
0. INPUT (ratings data) and c
(for trust on comparisons)
1. Compute from
2. Discard entries with fewer than
c comparisons
3. Set to be indices and
values of what’s left
4. = SVP( )
5. OUTPUT
31/40
Purdue ML Seminar
David Gleich, Purdue
32. Item Response Model
Synthetic evaluation
The synthetic results came from a model inspired by Ho and
Quinn [2008].
- center rating for user $i$
- sensitivity of user $i$
- value of item $j$
- error level in ratings
Sample ratings uniformly at random such that there
for expected ratings per user.
32/40
David F. Gleich (Purdue) Purdue
KDD 2011 ML Seminar
David Gleich, Purdue
21/20
33. Evaluation
Nuclear norm ranking
Mean rating
1 1
Median Kendall’s Tau
Median Kendall’s Tau
0.9 0.9
0.8 0.8
20
0.7 0.7
10
5
0.6 2 0.6
1.5
0.5 0.5
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Error Error
33/40
Figure 3: The performance of our algorithmPurdue
Purdue ML Seminar
David Gleich,
(left)
34. Conclusions and Future Work
Our motto
“aggregate, then complete”
1. Additional comparison
Rank aggregation with " 2. Noisy recovery! More
the nuclear norm is
realistic sampling.
principled
3. Skew-symmetric Lanczos
based SVD?
easy to compute
The results are much better than
simple approaches.
34/40
Purdue ML Seminar
David Gleich, Purdue