<!DOCTYPEHTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type"
content="text/html; charset=ISO-8859-1">
<title>AboutHap</title>
</head>
<body style="color: rgb(0, 0, 153); background-color: rgb(204, 255, 255);" link="#000066" vlink="#000066" alink="#000066">
<br>
<table style="text-align: left; margin-left: auto; margin-right: auto; color: rgb(0, 0, 102);"
border="0" cellpadding="20" cellspacing="10">
<tbody>
<tr align="center">
<thstyle="vertical-align: top;">
<tablestyle="width: 100%; text-align: left;" cellpadding="2"
cellspacing="2">
<tbody>
<tr>
<tdstyle="vertical-align: top;"><a href="aboutCubical.html"><small style="color: rgb(0, 0, 102);">Previous</small></a><br>
</td>
<td style="text-align: center; vertical-align: top; color: rgb(0, 0, 102);"><big><span style="font-weight: bold;">About HAP: Random Simplicial Complexes<br>
</span></big></td>
<tdstyle="text-align: right; vertical-align: top;"><a
href="aboutLinks.html"><smallstyle="color: rgb(0, 0, 102);">next</small></a><br>
</td>
</tr>
</tbody>
</table>
<big><spanstyle="font-weight: bold;"></span></big><br>
</th>
</tr>
<tr align="center">
<td style="vertical-align: top; background-color: rgb(255, 255, 255);"><big style="font-weight: bold;">Homology of random graphs and their nerves</big><br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 255); text-align: left;">For
a
positive
integer
n
and
probability
p
we
denote
by G(n,p) the
Erdos–Renyi random graph. This has n vertices with each potential edge
included independently with probability p. <br>
<br>
The following commands construct low dimensions of the<span style="font-style: italic;"> simlicial nerve</span> NG(n,p) of the
random graph on n=100 vertices for probability values in the range
0<p<0.5. By definition this nerve has n vertices and one
k-simplex for each collection of k+1 vertices spanning a complete
subgraph of G(n,p). <br>
<br>
Finally, for k=0,1,2,3,4 , the integral homology H<sub>k</sub>(NG(n,p),Z)
is
computed
and
stored
in
a
list
H.
<br>
</td>
</tr>
<tr>
<td style="background-color: rgb(255, 255, 204); vertical-align: top;">H:=[];;<br>
n:=100;;<br>
<br>
for p in [1..50]*(1/100) do<br>
K:=RandomSimplicialGraph(n,p);;<br>
G:=GraphOfSimplicialComplex(K);;<br>
N:=SimplicialNerveOfGraph(G,5);;<br>
Y:=SimplicialComplexToRegularCWComplex(N);;<br>
CriticalCellsOfRegularCWComplex(Y);;<br>
Add(H, List([0..4],i->Homology(Y,i)) );<br>
od;<br>
</td>
</tr>
<tr>
<td style="background-color: rgb(255, 255, 255); vertical-align: top;">The
following command shows that no torsion occurs in any of the computed
homology groups H<sub>k</sub>(NG(n,p),Z) . </td>
</tr>
<tr>
<td style="background-color: rgb(255, 255, 204); vertical-align: top;">gap>
SSortedList(Flat(H));<br>
[ 0 ]<br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 255);">It
is shown in [ M. Kahle and E. Meckes, "Limit theorems for Betti numbers
of random simplicial complexes", HHA, 15(1), 2013, 343-374] that the
Betti numbers <br>
<br>
<divstyle="text-align: center;">b<sub>k</sub>(n,p)= rank( H<sub>k</sub>(NG(n,p),Z)
)
<br>
</div>
<br>
are normally distributed for large n. The distribution is illustrated
by the following commands which plot b<sub>k</sub>(n,p)
against p for fixed n=100 and k=0,1,2,3,4.<br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 204);">gap>
B:=List([1..5],k->List(List(H,h->h[k]),x->Length(x)));;<br>
<br>
gap> for k in [0,1,2,3] do<br>
> A:=NullMat(Maximum(B[k+1])+5,100);;<br>
> for P in [1..50] do<br>
> A[B[k+1][P]+1][2*P]:=1;<br>
> od;<br>
>
M:=ThickenedPureCubicalComplex(PureCubicalComplex(TransposedMat(Reversed(A))));;<br>
> ViewPureCubicalComplex(M);<br>
> od;<br>
<br>
<divstyle="text-align: center;"><img style="width: 104px; height: 25px;" alt="" src="randomgraph1.png">
<imgstyle="width: 104px; height: 263px;" alt=""
src="randomgraph2.png"> <img style="width: 104px; height: 620px;" alt="" src="randomgraph3.png">
<img style="width: 104px; height: 1310px;" alt="" src="randomgraph4.png">
<imgstyle="width: 104px; height: 1072px;" alt=""
src="randomgraph5.png"> <br>
</div>
</td>
</tr>
<tr align="center">
<td style="vertical-align: top; background-color: rgb(255, 255, 255);"><big style="font-weight: bold;">Homology of random simplicial 2-complexes</big><br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 255);">For
a positive integer n and probability p we denote by Y(n,p) the
Linial-Meshulam random simplicial 2-complex. Its 1-skeleton is the
complete graph on n vertices; each possible 2-simplex is included
independently with probability p. <br>
<br>
The following commands compute the Betti numbers<br>
<br>
<divstyle="text-align: center;">rank( H<sub>1</sub>(Y(n,p),Z)
) <br>
</div>
<br>
for n=200 vertices and for p= 1/200, 2/200, 3/200, ... , 50/200. <br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 204);">L:=[];<br>
n:=200;;<br>
<br>
for p in [1..50]*(1/200) do<br>
K:=RandomSimplicialTwoComplex(n,p);<br>
H:=Homology(K,1);<br>
Add(L,H);<br>
od;<br>
<br>
List(L,x->Length(x));<br>
[ 13065, 6597, 1299, 393, 138, 48, 16, 7, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0 ]<br>
<br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 255);">The
following additional command shows that there is no torsion in any
of
the computed
homology groups H<sub>1</sub>(Y(n,p),Z) . <br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 204);">gap>
SSortedList(Flat(L));<br>
[ 0 ]<br>
</td>
</tr>
<tr align="center">
<td style="vertical-align: top; background-color: rgb(255, 255, 255);"><big style="font-weight: bold;">Fundamental groups of random simplicial
2-complexes</big><br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 255);">The
following commands compute presentations for the fundamental groups of
Y(n,p) for n=60 vertices and probabilities in the range 0<p<0.25.
<br>
<br>
The commands also list the number of generators for each fundamental
group, and show that in most cases there are no relators between
generators. Thus, in most (and possibly all) cases, the fundamental
group is free; for large p the fundamental group is trivial <br>
</td>
</tr>
<tr>
<td style="vertical-align: top; background-color: rgb(255, 255, 204);">gap>
L:=[];;<br>
gap> n:=60;;<br>
gap> for p in [1..50]*(1/200) do<br>
> K:=RandomSimplicialTwoComplex(n,p);;<br>
> Add(L,FundamentalGroup(K));<br>
> od;<br>
<br>
gap> NumbersOfGenerators:=List(L,g->Length(GeneratorsOfGroup(g)));<br>
[ 1542, 1357, 1206, 1024, 822, 647, 502, 344, 195, 110, 82, 57, 39, 18,
9, 14, 16, 15, 2, 4, 3, 2, 2, 2, 0, 0, 2, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]<br>
<br>
gap> NumbersOfRelations:=List(L,g->Length(RelatorsOfFpGroup(g)));<br>
[ 0, 0, 0, 0, 0, 0, 0, 0, 37, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0 ]<br>
</td>
</tr>
<tr>
<tdstyle="vertical-align: top;">
<table style="margin-left: auto; margin-right: auto; width: 100%; text-align: left;"
border="0" cellpadding="2" cellspacing="2">
<tbody>
<tr>
<tdstyle="vertical-align: top;"><a style="color: rgb(0, 0, 102);" href="aboutCubical.html">Previous
Page</a><br>
</td>
<tdstyle="text-align: center; vertical-align: top;"><a
href="aboutContents.html"><spanstyle="color: rgb(0, 0, 102);">Contents</span></a><br>
</td>
<tdstyle="text-align: right; vertical-align: top;"><a
href="aboutLinks.html"><spanstyle="color: rgb(0, 0, 102);">Next
page</span><br>
</a> </td>
</tr>
</tbody>
</table>
<a href="aboutTopology.html"><br>
</a> </td>
</tr>
</tbody>
</table>
<br>
<br>
</body>
</html>
Messung V0.5
¤ Dauer der Verarbeitung: 0.15 Sekunden
(vorverarbeitet)
¤
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.