SlideShare a Scribd company logo
1 of 18
Download to read offline
Phase Transitions of Best-of-Two and
Best-of-Three on Stochastic Block Models
Nobutaka Shimizu and Takeharu Shiraga
(Chuo University)(The University of Tokyo)
DISC19
/182
Sync. voting processes on a graph, Best-of-two (Bo2)
and Best-of-three (Bo3). Each vertex supports one of
two distinct opinions.
‣In each round, every node updates his opinion according
to some rule.
u
• (Bo2): Sample two random neighbors and . Then, take
the majority among opinions of and .
x y
x, y u
• (Bo3): Sample three random neighbors and take their majority.
‣Markov Chain on . Each state is denoted by .
‣ = consensus time starting from (r.v.).
2V
A ⊆ V
T(A) A
• c.f. In the well-known Voter process, adopts the opinion of a random neighbor.u
u y
x
u y
x
/18
Bo2 on with the half-and-half initial state.K500
3
/18
Known results (Bo2 & Bo3)
4
‣ For Bo2 and Bo3 on , w.h.p for any [DGMSS11].Kn T(A) = O(log n) A
‣ opinions setting: Bo2 and Bo3 on satisfies
for any if is moderate [BCNPT18], [GL18],
[BCNPST14].
k > 2 Kn
T(A) = O(k log n) A k
‣ For Bo2 and Bo3 on expander graphs, under
some initial bias assumption on [CER14], [CERRS15], [CRRS17].
T(A) = O(log n)
A
• core-periphery graph [CNNS18]
• regular stochastic block model [CNS19]
‣ Bo2 on various graphs are investigated recently
c.f. In Voter, it is known that [CEOR13].E[T(A)] = O(n)
/18
Known results (Bo2 & Bo3)
5
‣ w.h.p for any on [DGMSS11]. Much faster
than Voter.
T(A) = O(log n) A Kn
‣ If there are opinions, for any on if
is moderate [BCNPT18], [GL18], [BCNPST14].
k > 2 T(A) = O(k log n) A Kn
k
‣ Bo2 and Bo3 reaches consensus assumng an initial bias
on expander graphs [CER14], [CERRS15], [CRRS17]
O(log n)
• core-periphery graph [CNNS18]
• regular stochastic block model [CNS19]
‣ Bo2 on various graphs are investigated recently
c.f. In Voter, it is known that [CEOR13].E[T(A)] = O(n)
How graph structure affect the Bo2 and Bo3 ?
/18
This Work
‣Bo2 and Bo3 on Stochastic Block Model (SBM)
‣SBM is a random graph consists of two
vertex groups each of size , and two vertices
form an edge with probability
G(2n, p, q)
n
u, v
G(300, 0.1, 0.01)
6
if and are in the same group.u v
{
p
q o.w.
‣Generate and then set initial
opinion state and observe the process.
G(2n, p, q)
/187
Question: Where is the boundary?🤔?
q
p
= 0.4
q
p
= 0.1
/18
Our result (phase transition)
Consider Bo2 on such that and are constants.G(2n, p, q) p q
Theorem 1
• If , w.h.p. for any .
• If , w.h.p. for some .
q/p > 5 − 2 T(A) = O(log n) A ⊆ V
q/p < 5 − 2 T(A) ≥ exp(Ω(n)) A ⊆ V
8
Remarks:
1. Formally, we show that, w.h.p. exhibits a “nice”
structure such that the statements above hold.
G(2n, p, q)
2. Here, w.h.p. means “with probability for some const ”.1 − n−c
c > 0
3. Bo3 is more likely to reach consensus (since ).1/7 < 5 − 2
Consider Bo3 on such that and are constants.G(2n, p, q) p q
Theorem 2
• If , w.h.p. for any .
• If , w.h.p. for some .
q/p > 1/7 T(A) = O(log n) A ⊆ V
q/p < 1/7 T(A) ≥ exp(Ω(n)) A ⊆ V
/18
Two components of our proof
9
1. Structual property of .G(2n, p, q)
Tool: Janson’s ineq & Kim-Vu’s ineq
2. Bo2 and Bo3 on “nice” graphs
Tools: Jacobi matrix, competitive dynamical system.
/1810
Idea behind the proof (Bo3)
The case of (i.e., ) is previously known.
We expect that the same technique works for the case
of contsants .
p = q = 1 K2n
p, q > 0
‣Bo3 is a M.C. on in general graphs. But on in .
‣Let and be the set of opinion 1 and that in the next round.
Then
2V
{0,…, n} Kn
A A′
|A′| =
∑
v∈V
1[v∈A′]
is the sum of independent random variables, which concentrates
on its expectation.
E[|A′|] = ?
/19
α′ ≈ E[α′] =
E[|A′|]
n
= g(α) + O(n−1
) .
E[|A′| |A] =
∑
v∈A
Pr[v ∈ A′] =
∑
v∈V
g
(
degA(v)
deg(v) )
.
11
‣Evaluate . Indeed, letting
, we have
E[|A′|]
g(x) =
(
3
2)
x2
(1 − x) +
(
3
3)
x3
0
1
0.5 α g(α)
degA(v) = 3
v
‣Let On , we have
and thus
α = |A|/n . Kn
degA(v)
deg(v)
=
|A|
n
+ O(n−1
) = α + O(n−1
)
‣The orbit of is approximated by a discrete-time
dynamical system on .
α
[0,1]
deg(v) = 5
/18
Key lemma (concentration of for Bo3 on )E[α′|A] G(2n, p, p)
‣The lemma says “the behavior of can be approximated by a
discrete-time dynamical system.”
‣We can evaluate , which leads to the symmetry breaking.
‣The result can be extended to .
‣The proof relies on concentration inequalities.
α
Var[α′|A]
G(2n, p, q)
12
Suppose that . Then satisfies the
following w.h.p.:
p = ω(log n/n) G(2n, p, p)
∀A ⊆ V, E[α′|A] = g(α) +O ((np)−0.5
),
where g(x) := 3x2
− 2x3
.
‣This observation holds for . We prove it for !!Kn G(2n, p, q)
/18
On G(2n,p,q)
‣ Let .Ai := A ∩ Vi (i = 1,2)
A1
A2
13
‣ Fix the time step .
‣ : configuration at the current step .
‣ : that of the next step .
t
A t
A′ t + 1
/18
2D dynamical system
‣Consider for each .E[α′i] = E[|A′i |/n] i = 1,2
• This yields a 2D dynamical system on .[0,1]2
14
α′1 = 3 (
α1 + rα2
1 + r )
2
− 2 (
α1 + rα2
1 + r )
3
± O((np)−1/2
),
α′2 = 3 (
rα1 + α2
1 + r )
2
− 2 (
rα1 + α2
1 + r )
3
± O((np)−1/2
) .
‣For Bo3, our key lemma implies
where r = q/p .
/18
thresholdr <
15
thresholdr >
Vertical axis = Horisontal axis = .α1 α2
/1816
threshold (absorbing)r < threshold (saddle).r >
/18
Intuitive Explanation (p≠q)
‣If , two absorbing points appear:q/p < 1/7
It takes exponential time to escape from the ε-
neighborhood of the absorbing points.
17
‣We used competitive dynamical system theory to prove
the orbit convergence for any initial point .(α1, α2)
‣We do not explore the case where .r = 1/7
‣If , these absorbing points disappear:q/p > 1/7
Escape from the fixed point in time.O(log n)
/18
• Approximate orbit by a dynamical system.
Summary
‣ Technical tools:
• Local behavior around fixed points.
• Global behavior (orbit convergence)
‣ Bo2 and Bo3 on SBM
• phase transition on q/p.
• consensus time: log vs. exponential.
18
• Community detection is a potential application.
• The case threshold is left as future task.
• We obtained similar results for sparse SBM under initial bias
assumption.
q/p =
Concentration inequalities to analyze random graphs.
Essentially known in [Doerr et al. 2011].
Competitive dynamical system theory.

More Related Content

What's hot

Gaussian Quadrature Formula
Gaussian Quadrature FormulaGaussian Quadrature Formula
Gaussian Quadrature FormulaDhaval Shukla
 
Gauss Quadrature Formula
Gauss Quadrature FormulaGauss Quadrature Formula
Gauss Quadrature FormulaMaitree Patel
 
Even Harmonious Labeling of the Graph H (2n, 2t+1)
Even Harmonious Labeling of the Graph H (2n, 2t+1)Even Harmonious Labeling of the Graph H (2n, 2t+1)
Even Harmonious Labeling of the Graph H (2n, 2t+1)inventionjournals
 
Unit 2 analysis of continuous time signals-mcq questions
Unit 2   analysis of continuous time signals-mcq questionsUnit 2   analysis of continuous time signals-mcq questions
Unit 2 analysis of continuous time signals-mcq questionsDr.SHANTHI K.G
 
Theory of Automata and formal languages Unit 3
Theory of Automata and formal languages Unit 3Theory of Automata and formal languages Unit 3
Theory of Automata and formal languages Unit 3Abhimanyu Mishra
 
Further pure mathmatics 3 vectors
Further pure mathmatics 3 vectorsFurther pure mathmatics 3 vectors
Further pure mathmatics 3 vectorsDennis Almeida
 
Numerical integration;Gaussian integration one point, two point and three poi...
Numerical integration;Gaussian integration one point, two point and three poi...Numerical integration;Gaussian integration one point, two point and three poi...
Numerical integration;Gaussian integration one point, two point and three poi...vaibhav tailor
 
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULA
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULANUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULA
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULAKHORASIYA DEVANSU
 
2.4 mst kruskal’s
2.4 mst  kruskal’s 2.4 mst  kruskal’s
2.4 mst kruskal’s Krish_ver2
 
Graph Algorithms: Breadth-First Search (BFS)
Graph Algorithms: Breadth-First Search (BFS)Graph Algorithms: Breadth-First Search (BFS)
Graph Algorithms: Breadth-First Search (BFS)Md. Shafiuzzaman Hira
 
Bellmanfordwith negative cycle js
Bellmanfordwith negative cycle jsBellmanfordwith negative cycle js
Bellmanfordwith negative cycle jsJeeSa Sultana
 
BFS, Breadth first search | Search Traversal Algorithm
BFS, Breadth first search | Search Traversal AlgorithmBFS, Breadth first search | Search Traversal Algorithm
BFS, Breadth first search | Search Traversal AlgorithmMSA Technosoft
 
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...inventionjournals
 
B.tech admission in india
B.tech admission in indiaB.tech admission in india
B.tech admission in indiaEdhole.com
 
Clustering in Hilbert geometry for machine learning
Clustering in Hilbert geometry for machine learningClustering in Hilbert geometry for machine learning
Clustering in Hilbert geometry for machine learningFrank Nielsen
 
2021 preTEST4A Vector Calculus
2021 preTEST4A Vector Calculus2021 preTEST4A Vector Calculus
2021 preTEST4A Vector CalculusA Jorge Garcia
 

What's hot (20)

Quadrature
QuadratureQuadrature
Quadrature
 
Hw2
Hw2Hw2
Hw2
 
Gaussian Quadrature Formula
Gaussian Quadrature FormulaGaussian Quadrature Formula
Gaussian Quadrature Formula
 
Gauss Quadrature Formula
Gauss Quadrature FormulaGauss Quadrature Formula
Gauss Quadrature Formula
 
Even Harmonious Labeling of the Graph H (2n, 2t+1)
Even Harmonious Labeling of the Graph H (2n, 2t+1)Even Harmonious Labeling of the Graph H (2n, 2t+1)
Even Harmonious Labeling of the Graph H (2n, 2t+1)
 
Unit 2 analysis of continuous time signals-mcq questions
Unit 2   analysis of continuous time signals-mcq questionsUnit 2   analysis of continuous time signals-mcq questions
Unit 2 analysis of continuous time signals-mcq questions
 
Theory of Automata and formal languages Unit 3
Theory of Automata and formal languages Unit 3Theory of Automata and formal languages Unit 3
Theory of Automata and formal languages Unit 3
 
Further pure mathmatics 3 vectors
Further pure mathmatics 3 vectorsFurther pure mathmatics 3 vectors
Further pure mathmatics 3 vectors
 
Numerical integration;Gaussian integration one point, two point and three poi...
Numerical integration;Gaussian integration one point, two point and three poi...Numerical integration;Gaussian integration one point, two point and three poi...
Numerical integration;Gaussian integration one point, two point and three poi...
 
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULA
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULANUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULA
NUMERICAL INTEGRATION : ERROR FORMULA, GAUSSIAN QUADRATURE FORMULA
 
2.4 mst kruskal’s
2.4 mst  kruskal’s 2.4 mst  kruskal’s
2.4 mst kruskal’s
 
Graph Algorithms: Breadth-First Search (BFS)
Graph Algorithms: Breadth-First Search (BFS)Graph Algorithms: Breadth-First Search (BFS)
Graph Algorithms: Breadth-First Search (BFS)
 
Gch12 l2
Gch12 l2Gch12 l2
Gch12 l2
 
Bellmanfordwith negative cycle js
Bellmanfordwith negative cycle jsBellmanfordwith negative cycle js
Bellmanfordwith negative cycle js
 
BFS, Breadth first search | Search Traversal Algorithm
BFS, Breadth first search | Search Traversal AlgorithmBFS, Breadth first search | Search Traversal Algorithm
BFS, Breadth first search | Search Traversal Algorithm
 
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...
Establishment of New Special Deductions from Gauss Divergence Theorem in a Ve...
 
B.tech admission in india
B.tech admission in indiaB.tech admission in india
B.tech admission in india
 
Clustering in Hilbert geometry for machine learning
Clustering in Hilbert geometry for machine learningClustering in Hilbert geometry for machine learning
Clustering in Hilbert geometry for machine learning
 
Optimisation random graph presentation
Optimisation random graph presentationOptimisation random graph presentation
Optimisation random graph presentation
 
2021 preTEST4A Vector Calculus
2021 preTEST4A Vector Calculus2021 preTEST4A Vector Calculus
2021 preTEST4A Vector Calculus
 

Similar to Phase Transitions of Best-of-Two and Best-of-Three on Stochastic Block Models

Solucao_Marion_Thornton_Dinamica_Classic (1).pdf
Solucao_Marion_Thornton_Dinamica_Classic (1).pdfSolucao_Marion_Thornton_Dinamica_Classic (1).pdf
Solucao_Marion_Thornton_Dinamica_Classic (1).pdfFranciscoJavierCaedo
 
Phase diagram at finite T & Mu in strong coupling limit of lattice QCD
Phase diagram at finite T & Mu in strong coupling limit of lattice QCDPhase diagram at finite T & Mu in strong coupling limit of lattice QCD
Phase diagram at finite T & Mu in strong coupling limit of lattice QCDBenjamin Jaedon Choi
 
Triumph- JEE Advanced Maths - Paper 1
Triumph- JEE Advanced Maths - Paper 1Triumph- JEE Advanced Maths - Paper 1
Triumph- JEE Advanced Maths - Paper 1askiitians
 
Cpt 5 synthesis of linkages
Cpt 5 synthesis of linkagesCpt 5 synthesis of linkages
Cpt 5 synthesis of linkagesMohit Jain
 
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...Alpro
 
Notes and formulae mathematics
Notes and formulae mathematicsNotes and formulae mathematics
Notes and formulae mathematicsZainonie Ma'arof
 
Multiband Transceivers - [Chapter 1]
Multiband Transceivers - [Chapter 1] Multiband Transceivers - [Chapter 1]
Multiband Transceivers - [Chapter 1] Simen Li
 
2014 04 22 wits presentation oqw
2014 04 22 wits presentation oqw2014 04 22 wits presentation oqw
2014 04 22 wits presentation oqwRene Kotze
 
Lecture -Earthquake Engineering (2).pdf
Lecture -Earthquake Engineering (2).pdfLecture -Earthquake Engineering (2).pdf
Lecture -Earthquake Engineering (2).pdfRayRabara
 
Stochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersStochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersTaiji Suzuki
 
25 johnarry tonye ngoji 250-263
25 johnarry tonye ngoji 250-26325 johnarry tonye ngoji 250-263
25 johnarry tonye ngoji 250-263Alexander Decker
 
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...Qualitative measurement of Klauder coherent states using Bohmian machanics, C...
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...Sanjib Dey
 
State Space Realizations_new.pptx
State Space Realizations_new.pptxState Space Realizations_new.pptx
State Space Realizations_new.pptxMohdNajibAliMokhtar
 
New Families of Odd Harmonious Graphs
New Families of Odd Harmonious GraphsNew Families of Odd Harmonious Graphs
New Families of Odd Harmonious GraphsIJMIT JOURNAL
 
Introduction to Bayesian Inference
Introduction to Bayesian InferenceIntroduction to Bayesian Inference
Introduction to Bayesian InferencePeter Chapman
 

Similar to Phase Transitions of Best-of-Two and Best-of-Three on Stochastic Block Models (20)

Solucao_Marion_Thornton_Dinamica_Classic (1).pdf
Solucao_Marion_Thornton_Dinamica_Classic (1).pdfSolucao_Marion_Thornton_Dinamica_Classic (1).pdf
Solucao_Marion_Thornton_Dinamica_Classic (1).pdf
 
Phase diagram at finite T & Mu in strong coupling limit of lattice QCD
Phase diagram at finite T & Mu in strong coupling limit of lattice QCDPhase diagram at finite T & Mu in strong coupling limit of lattice QCD
Phase diagram at finite T & Mu in strong coupling limit of lattice QCD
 
Triumph- JEE Advanced Maths - Paper 1
Triumph- JEE Advanced Maths - Paper 1Triumph- JEE Advanced Maths - Paper 1
Triumph- JEE Advanced Maths - Paper 1
 
Cpt 5 synthesis of linkages
Cpt 5 synthesis of linkagesCpt 5 synthesis of linkages
Cpt 5 synthesis of linkages
 
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...
Mathematics HL and Further mathematics HL Formula Booklet Zimsec Cambridge Zi...
 
Notes and formulae mathematics
Notes and formulae mathematicsNotes and formulae mathematics
Notes and formulae mathematics
 
Multiband Transceivers - [Chapter 1]
Multiband Transceivers - [Chapter 1] Multiband Transceivers - [Chapter 1]
Multiband Transceivers - [Chapter 1]
 
3D Geometry Theory 3
3D Geometry Theory 33D Geometry Theory 3
3D Geometry Theory 3
 
2014 04 22 wits presentation oqw
2014 04 22 wits presentation oqw2014 04 22 wits presentation oqw
2014 04 22 wits presentation oqw
 
Mathematics
MathematicsMathematics
Mathematics
 
Lecture -Earthquake Engineering (2).pdf
Lecture -Earthquake Engineering (2).pdfLecture -Earthquake Engineering (2).pdf
Lecture -Earthquake Engineering (2).pdf
 
Stochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersStochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of Multipliers
 
25 johnarry tonye ngoji 250-263
25 johnarry tonye ngoji 250-26325 johnarry tonye ngoji 250-263
25 johnarry tonye ngoji 250-263
 
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...Qualitative measurement of Klauder coherent states using Bohmian machanics, C...
Qualitative measurement of Klauder coherent states using Bohmian machanics, C...
 
State Space Realizations_new.pptx
State Space Realizations_new.pptxState Space Realizations_new.pptx
State Space Realizations_new.pptx
 
ME Reference.pdf
ME Reference.pdfME Reference.pdf
ME Reference.pdf
 
New Families of Odd Harmonious Graphs
New Families of Odd Harmonious GraphsNew Families of Odd Harmonious Graphs
New Families of Odd Harmonious Graphs
 
Escola naval 2015
Escola naval 2015Escola naval 2015
Escola naval 2015
 
Introduction to Bayesian Inference
Introduction to Bayesian InferenceIntroduction to Bayesian Inference
Introduction to Bayesian Inference
 
Beam buckling
Beam bucklingBeam buckling
Beam buckling
 

Recently uploaded

Module for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learningModule for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learninglevieagacer
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfrohankumarsinghrore1
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryAlex Henderson
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bSérgio Sacani
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPirithiRaju
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformationAreesha Ahmad
 
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICESAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICEayushi9330
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Joonhun Lee
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑Damini Dixit
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)Areesha Ahmad
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPirithiRaju
 
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...Monika Rani
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)Areesha Ahmad
 
Formation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksFormation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksSérgio Sacani
 
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 60009654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000Sapana Sha
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPirithiRaju
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)Areesha Ahmad
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLkantirani197
 

Recently uploaded (20)

Module for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learningModule for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learning
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdf
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformation
 
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICESAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdf
 
CELL -Structural and Functional unit of life.pdf
CELL -Structural and Functional unit of life.pdfCELL -Structural and Functional unit of life.pdf
CELL -Structural and Functional unit of life.pdf
 
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)
 
Formation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksFormation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disks
 
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 60009654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
 

Phase Transitions of Best-of-Two and Best-of-Three on Stochastic Block Models

  • 1. Phase Transitions of Best-of-Two and Best-of-Three on Stochastic Block Models Nobutaka Shimizu and Takeharu Shiraga (Chuo University)(The University of Tokyo) DISC19
  • 2. /182 Sync. voting processes on a graph, Best-of-two (Bo2) and Best-of-three (Bo3). Each vertex supports one of two distinct opinions. ‣In each round, every node updates his opinion according to some rule. u • (Bo2): Sample two random neighbors and . Then, take the majority among opinions of and . x y x, y u • (Bo3): Sample three random neighbors and take their majority. ‣Markov Chain on . Each state is denoted by . ‣ = consensus time starting from (r.v.). 2V A ⊆ V T(A) A • c.f. In the well-known Voter process, adopts the opinion of a random neighbor.u u y x u y x
  • 3. /18 Bo2 on with the half-and-half initial state.K500 3
  • 4. /18 Known results (Bo2 & Bo3) 4 ‣ For Bo2 and Bo3 on , w.h.p for any [DGMSS11].Kn T(A) = O(log n) A ‣ opinions setting: Bo2 and Bo3 on satisfies for any if is moderate [BCNPT18], [GL18], [BCNPST14]. k > 2 Kn T(A) = O(k log n) A k ‣ For Bo2 and Bo3 on expander graphs, under some initial bias assumption on [CER14], [CERRS15], [CRRS17]. T(A) = O(log n) A • core-periphery graph [CNNS18] • regular stochastic block model [CNS19] ‣ Bo2 on various graphs are investigated recently c.f. In Voter, it is known that [CEOR13].E[T(A)] = O(n)
  • 5. /18 Known results (Bo2 & Bo3) 5 ‣ w.h.p for any on [DGMSS11]. Much faster than Voter. T(A) = O(log n) A Kn ‣ If there are opinions, for any on if is moderate [BCNPT18], [GL18], [BCNPST14]. k > 2 T(A) = O(k log n) A Kn k ‣ Bo2 and Bo3 reaches consensus assumng an initial bias on expander graphs [CER14], [CERRS15], [CRRS17] O(log n) • core-periphery graph [CNNS18] • regular stochastic block model [CNS19] ‣ Bo2 on various graphs are investigated recently c.f. In Voter, it is known that [CEOR13].E[T(A)] = O(n) How graph structure affect the Bo2 and Bo3 ?
  • 6. /18 This Work ‣Bo2 and Bo3 on Stochastic Block Model (SBM) ‣SBM is a random graph consists of two vertex groups each of size , and two vertices form an edge with probability G(2n, p, q) n u, v G(300, 0.1, 0.01) 6 if and are in the same group.u v { p q o.w. ‣Generate and then set initial opinion state and observe the process. G(2n, p, q)
  • 7. /187 Question: Where is the boundary?🤔? q p = 0.4 q p = 0.1
  • 8. /18 Our result (phase transition) Consider Bo2 on such that and are constants.G(2n, p, q) p q Theorem 1 • If , w.h.p. for any . • If , w.h.p. for some . q/p > 5 − 2 T(A) = O(log n) A ⊆ V q/p < 5 − 2 T(A) ≥ exp(Ω(n)) A ⊆ V 8 Remarks: 1. Formally, we show that, w.h.p. exhibits a “nice” structure such that the statements above hold. G(2n, p, q) 2. Here, w.h.p. means “with probability for some const ”.1 − n−c c > 0 3. Bo3 is more likely to reach consensus (since ).1/7 < 5 − 2 Consider Bo3 on such that and are constants.G(2n, p, q) p q Theorem 2 • If , w.h.p. for any . • If , w.h.p. for some . q/p > 1/7 T(A) = O(log n) A ⊆ V q/p < 1/7 T(A) ≥ exp(Ω(n)) A ⊆ V
  • 9. /18 Two components of our proof 9 1. Structual property of .G(2n, p, q) Tool: Janson’s ineq & Kim-Vu’s ineq 2. Bo2 and Bo3 on “nice” graphs Tools: Jacobi matrix, competitive dynamical system.
  • 10. /1810 Idea behind the proof (Bo3) The case of (i.e., ) is previously known. We expect that the same technique works for the case of contsants . p = q = 1 K2n p, q > 0 ‣Bo3 is a M.C. on in general graphs. But on in . ‣Let and be the set of opinion 1 and that in the next round. Then 2V {0,…, n} Kn A A′ |A′| = ∑ v∈V 1[v∈A′] is the sum of independent random variables, which concentrates on its expectation. E[|A′|] = ?
  • 11. /19 α′ ≈ E[α′] = E[|A′|] n = g(α) + O(n−1 ) . E[|A′| |A] = ∑ v∈A Pr[v ∈ A′] = ∑ v∈V g ( degA(v) deg(v) ) . 11 ‣Evaluate . Indeed, letting , we have E[|A′|] g(x) = ( 3 2) x2 (1 − x) + ( 3 3) x3 0 1 0.5 α g(α) degA(v) = 3 v ‣Let On , we have and thus α = |A|/n . Kn degA(v) deg(v) = |A| n + O(n−1 ) = α + O(n−1 ) ‣The orbit of is approximated by a discrete-time dynamical system on . α [0,1] deg(v) = 5
  • 12. /18 Key lemma (concentration of for Bo3 on )E[α′|A] G(2n, p, p) ‣The lemma says “the behavior of can be approximated by a discrete-time dynamical system.” ‣We can evaluate , which leads to the symmetry breaking. ‣The result can be extended to . ‣The proof relies on concentration inequalities. α Var[α′|A] G(2n, p, q) 12 Suppose that . Then satisfies the following w.h.p.: p = ω(log n/n) G(2n, p, p) ∀A ⊆ V, E[α′|A] = g(α) +O ((np)−0.5 ), where g(x) := 3x2 − 2x3 . ‣This observation holds for . We prove it for !!Kn G(2n, p, q)
  • 13. /18 On G(2n,p,q) ‣ Let .Ai := A ∩ Vi (i = 1,2) A1 A2 13 ‣ Fix the time step . ‣ : configuration at the current step . ‣ : that of the next step . t A t A′ t + 1
  • 14. /18 2D dynamical system ‣Consider for each .E[α′i] = E[|A′i |/n] i = 1,2 • This yields a 2D dynamical system on .[0,1]2 14 α′1 = 3 ( α1 + rα2 1 + r ) 2 − 2 ( α1 + rα2 1 + r ) 3 ± O((np)−1/2 ), α′2 = 3 ( rα1 + α2 1 + r ) 2 − 2 ( rα1 + α2 1 + r ) 3 ± O((np)−1/2 ) . ‣For Bo3, our key lemma implies where r = q/p .
  • 15. /18 thresholdr < 15 thresholdr > Vertical axis = Horisontal axis = .α1 α2
  • 16. /1816 threshold (absorbing)r < threshold (saddle).r >
  • 17. /18 Intuitive Explanation (p≠q) ‣If , two absorbing points appear:q/p < 1/7 It takes exponential time to escape from the ε- neighborhood of the absorbing points. 17 ‣We used competitive dynamical system theory to prove the orbit convergence for any initial point .(α1, α2) ‣We do not explore the case where .r = 1/7 ‣If , these absorbing points disappear:q/p > 1/7 Escape from the fixed point in time.O(log n)
  • 18. /18 • Approximate orbit by a dynamical system. Summary ‣ Technical tools: • Local behavior around fixed points. • Global behavior (orbit convergence) ‣ Bo2 and Bo3 on SBM • phase transition on q/p. • consensus time: log vs. exponential. 18 • Community detection is a potential application. • The case threshold is left as future task. • We obtained similar results for sparse SBM under initial bias assumption. q/p = Concentration inequalities to analyze random graphs. Essentially known in [Doerr et al. 2011]. Competitive dynamical system theory.