Matrix Martingales in Randomized Numerical Linear Algebra

Matrix Martingales in Randomized Numerical Linear Algebra

Speaker Rasmus Kyng Harvard University

Sep 27, 2018

Concentration of Scalar Random Variables

Random ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. (!$ = 0

Is ! ≈ 0 with high probability?

Concentration of Scalar Random Variables

Bernstein’s InequalityRandom ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. (!$ = 03. !$ ≤ +4. ∑$ (!$, ≤ -,

gives

ℙ[ ! > 1] ≤ 2exp − 1,/2+1 + -, ≤ :

E.g. if 1 = 0.5, and +, -, = 0.1/log(1/:)

Concentration of Scalar Martingales

Random ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. (!$ = 0 ( !$|!+, … , !$-+ = 0


Random ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. (!$ = 0 ( !$| previous steps = 0

Concentration of Scalar MartingalesFreedman’s InequalityBernstein’s InequalityRandom ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. (!$ = 03. !$ ≤ +4. ∑$ (!$, ≤ -,

gives

ℙ[ ! > 1] ≤ 2 exp − 1,/2+1 + -,

( !$| previous steps = 0

∑$ ( !$,| previous steps ≤ -,

+B

ℙ ∑$ ( !$,| previous steps > -, ≤ B


Freedman’s InequalityRandom ! = ∑$ !$ , !$ ∈ ℝ1. !$ are independent2. ( !$| previous steps = 03. !$ ≤ 54. ℙ[∑$ ( !$8| previous steps > :8] ≤ <

gives

ℙ[ ! > =] ≤ 2exp − =8/25= + :8 + <

Concentration of Matrix Random Variables

Matrix Bernstein’s Inequality (Tropp 11)Random ! = ∑$ !$ !$ ∈ ℝ'×', symmetric1. !$ are independent2. *!$ = +3. !$ ≤ /4. ∑$ *!$0 ≤ 10

gives

ℙ[ ! > 5] ≤ 7 2exp − =>/0@=AB>

Concentration of Matrix Martingales

Matrix Freedman’s Inequality (Tropp 11)Random ! = ∑$ !$ !$ ∈ ℝ'×', symmetric1. !$ are independent2. *!$ = 0 * !$| previous steps = 03. !$ ≤ 94. ∑$ *!$: ≤ ;: ℙ[ ∑$ * !$:| prev. steps > ;:] ≤ @

gives

ℙ[ ! > A] ≤ B 2exp − FG/:IFJKG

+ @

E.g. if A = 0.5, and 9, ;: = 0.1/log(B/R)

≤ R + @

Concentration of Matrix Martingales

Matrix Freedman’s Inequality (Tropp 11)

ℙ[ ∑$ % &$'| prev. steps > 1'] ≤ 4

Predictable quadratic variation =6

$% &$'| prev. steps

Laplacian Matrices

! "Graph # = (&, ()Edge weights *:( → ℝ./ = &0 = |(|

2

/×/ matrix 4 = 56∈8

46

Laplacian Matrices

! "Graph # = (&, ()Edge weights *:( → ℝ./ = &0 = |(|

2

3 = 45∈7

35

L(a,b) = w(a,b)

. . . a . . . b . . .�

��

�

��

...a 1 �1...b �1 1...

/×/ matrix

w(a,b)

Laplacian Matrices

!: weighted adjacency matrix of the graph

#: diagonal matrix of weighted degrees

$ = # − !

!'( = )'(

#'' = ∑( )'(

Graph + = (-, /)Edge weights ):/ → ℝ34 = -5 = |/|

Laplacian Matrices

Symmetric matrix !

All off-diagonals are non-positive and

!"" =$%&"

!"%

Laplacian of a Graph

�

�1 �1 0

�1 1 00 0 0

�

�

�

�0 0 00 1 �10 �1 1

�

�2�

�2 0 �20 0 0

�2 0 2

�

�

11

�

�3 �1 �2

�1 2 �1�2 �1 3

�

�

Laplacian Matrices

[ST04]: solving Laplacian linear equations in !" # time

⋮

[KS16]: simple algorithm

Solving a Laplacian Linear Equation

!" = $

Gaussian EliminationFind %, upper triangular matrix, s.t.

%&% = !Then

" = %'(%'&$

Easy to apply %'( and %'&

Solving a Laplacian Linear Equation

!" = $

Approximate Gaussian EliminationFind %, upper triangular matrix, s.t.

%&% ≈ !

% is sparse.

( log ,- iterations to get.-approximate solution /".

Approximate Gaussian Elimination

Theorem [KS]When ! is an Laplacian matrix with " non-zeros,we can find in #(" log( )) time an upper triangular matrix + with #(" log( )) non-zeros,s.t. w.h.p.

+,+ ≈ !

Additive View of Gaussian Elimination

Find !, upper triangular matrix, s.t !"! = $

0

BB@

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

1

CCA$ =

�

��

16 �4 �8 �4�4 1 2 1�8 2 4 2�4 1 2 1

�

��

�

��

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

�

��

�

��

16 �4 �8 �4�4 1 2 1�8 2 4 2�4 1 2 1

�

�� =

�

��

4�1�2�1

�

��

�

��

4�1�2�1

�

��

�

Find the rank-1 matrix that agrees with !on the first row and column.


�

��

16 �4 �8 �4�4 1 2 1�8 2 4 2�4 1 2 1

�

��

�

��

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

�

��

�

��

0 0 0 00 4 �2 �20 �2 10 �20 �2 �2 6

�

��

Subtract the rank 1 matrix.We have eliminated the first variable.

=


The remaining matrix is PSD.

�

��

0 0 0 00 4 �2 �20 �2 10 �20 �2 �2 6

�

��


�

��

0 0 0 00 4 �2 �20 �2 10 �20 �2 �2 6

�

��

Find rank-1 matrix that agrees withour matrix on the next row and column.

�

��

0 0 0 00 4 �2 �20 �2 1 10 �2 1 1

�

�� =

�

��

02

�1�1

�

��

�

��

02

�1�1

�

��

�


Subtract the rank 1 matrix.We have eliminated the second variable.

�

��

0 0 0 00 4 �2 �20 �2 10 �20 �2 �2 6

�

��

�

��

0 0 0 00 4 �2 �20 �2 1 10 �2 1 1

�

��

�

�

��

0 0 0 00 0 0 00 0 9 �30 0 �3 5

�

��=


0

BB@

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

1

CCA! =

Repeat until all parts written as rank 1 terms.

=

�

��

4�1�2�1

�

��

�

��

4�1�2�1

�

��

�

+

�

��

02

�1�1

�

��

�

��

02

�1�1

�

��

�

+

�

��

003

�1

�

��

�

��

003

�1

�

��

�

+

�

��

0002

�

��

�

��

0002

�

��

�


0

BB@

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

1

CCA

=

�

��

4 0 0 0�1 2 0 0�2 �1 3 0�1 �1 �1 2

�

��

�

��

4 �1 �2 �10 2 �1 �10 0 3 �10 0 0 2

�

��



! =

0

BB@

16 �4 �8 �4�4 5 0 �1�8 0 14 0�4 �1 0 7

1

CCA

=

�

��

4 �1 �2 �10 2 �1 �10 0 3 �10 0 0 2

�

��

� �

��

4 �1 �2 �10 2 �1 �10 0 3 �10 0 0 2

�

�� = "# "



$ =


What is special about Gaussian Elimination on Laplacians?

The remaining matrix is always Laplacian.

L =

�

��

16 �8 �4 �4�8 8 0 0�4 0 4 0�4 0 0 4

�

��! =




L =

�

��

16 �8 �4 �4�8 4 2 2�4 2 1 1�4 2 1 1

�

�� +

�

��

0 0 0 00 4 �2 �20 �2 3 �10 �2 �1 3

�

��! =




+

�

��

0 0 0 00 4 �2 �20 �2 3 �10 �2 �1 3

�

��

A new Laplacian!

4 �2 �2�2 3 �1�2 �1 3

L =

�

��

4�2�1�1

�

��

�

��

4�2�1�1

�

��

�

! =

Why is Gaussian Elimination Slow?

Solving !" = $ by Gaussian Elimination can take Ω &' time.

The main issue is fill

(! =




New Laplacian(

Elimination creates a cliqueon the neighbors of (

! =




(

Laplacian cliques can be sparsified!

New Laplacian

≈

! =

Gaussian Elimination

1. Pick a vertex ! to eliminate2. Add the clique created by eliminating !3. Repeat until done


1. Pick a vertex ! to eliminate2. Add the clique created by eliminating !3. Repeat until done


1. Pick a random vertex ! to eliminate2. Add the clique created by eliminating !3. Repeat until done


1. Pick a random vertex ! to eliminate2. Sample the clique created by eliminating !3. Repeat until done

Resembles randomized Incomplete Cholesky

Approximating Matrices by Sampling

Goal !"! ≈ $

Approach1. % !"! = $2. Show !"! concentrated around expectation

Gives !"! ≈ $ w. high probability

Approximating Matrices in Expectation

Consider eliminating the first variable

Original Laplacian

Rank 1term

Remaining graph + clique

! "($) = '( '()

+ ! ∎ ∎∎ ∎

∎∎

∎ ∎ ∎



Rank 1term

Remaining graph + sparsified clique

! "($) = '$ '$(

+ ! ∎ ∎∎ ∎

∎∎

∎ ∎ ∎



! "($) = '$ '$(

+ ! ∎ ∎∎ ∎

∎∎

∎ ∎ ∎Rank 1term




! "($) = '$ '$(

+ ! ∎ ∎∎ ∎

∎∎



! +($) = '$ '$(

+ ! ∎ ∎∎ ∎

∎∎

∎ ∎ ∎Suppose



! "($) = '$ '$(

+ ! ∎ ∎∎ ∎

∎∎



= "(+)Then


Let !(#) be our approximation after % eliminations

If we ensure at each step

Then& !(#) = !(#())

& ∎ ∎∎ ∎

∎∎

∎ ∎ ∎= ∎ ∎

∎ ∎∎∎

∎ ∎ ∎Sparsified clique Clique

& ! # −! #() | previous steps = 6

Approximation?

Approximate Gaussian EliminationFind !, upper triangular matrix, s.t.

!"! ≈ $

!"! − $ ≤ 0.5

$*+/-!"!$*+/- − . ≤ 0.5

Essential Tools

Isotropic position !" ≝ $%&/( " $%&/(

Goal is now)*) − , ≈ .

PSD Order/ ≼ 1

iff for all 22*/2 ≤ 2*12

Matrix Concentration: Edge Variables

Zero-mean variables !" = $" − &"

Isotropic position variables '!"

w. probability ("

o.w. $" =

1("&*

1("+

Predictable Quadratic Variation

Predictable quadratic variation = ∑# $ %#&| prev. steps

Want to show ℙ ∑# $ %#&| prev. steps > 1& ≤ 3

Promise: $4%5& ≼ 7895

Sample Variance

!" #$∈ "

&'$ ≼ !"#$∋"

&'$

≼

* *

elim. clique of

Sample Variance

!" #$∈ "

&'$ ≼ !"#$∋"

&'$

≼

* *

elim. clique of

≼ 4&'# vertices

= 4-# vertices

= 2current lap.# vertices

WARNING: only true w.h.p.

= " 4$

Sample Variance

Recall %&'() ≼ "+,(

Putting it together%- .(∈ -

+,( ≼ %-.(∋-

+,(elim. clique of

= 4$# vertices

%- .(∈ -

%&'() ≼elim. clique of # vertices

variance in one round of elimination

Sample Variance

! ≼rounds ofelimination

variance

≼ 4 $ log ( ⋅ *

! $ ⋅ 4*# ver2ces

rounds ofelimination

Summary

Matrix martingales: a natural fit for algorithmic analysis

Understanding the Predictable Quadratic Variation is key

Some results using matrix martingales

Cohen, Musco, Pachocki ’16 – online row sampling

Kyng, Sachdeva ’17 – approximate Gaussian elimination

Kyng, Peng, Pachocki, Sachdeva ’17 – semi-streaming graph sparsification

Cohen, Kelner, Kyng, Peebles, Peng, Rao, Sidford ’18 – solving Directed Laplacian eqs.

Kyng, Song ’18 – Matrix Chernoff bound for negatively dependent random variables

Thanks!

How to Sample a Clique

For each edge (1, $)pick an edge 1, & with probability ~ ()insert edge $, & with weight *+*,*+-*,

1

23

4

5 54

32

1

23

4

5 54

32



Practice

Julia Package: Laplacians.jl

tiny.cc/spielman-solver-code

Theory

Nearly-linear time Directed Laplacian solversusing approximate Gaussian elimination[CKKPPSR17]

This is really the end

Matrix Martingales in Randomized Numerical Linear Algebra

Documents

Transcript of Matrix Martingales in Randomized Numerical Linear Algebra