Logo: to the web site of Uppsala University

uu.sePublications from Uppsala University
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Decorating trees grown in urns
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Mathematics.
2022 (English)Doctoral thesis, comprehensive summary (Other academic)
Description
Abstract [en]

Random recursive trees are classic models of random trees. A random recursive tree is initiated with a single root vertex and constructed in steps, whereby at each step a vertex is added as the child of a vertex chosen uniformly at random in the tree. A preferential attachment tree is constructed in a similar manner, except the random choice of the vertex at each step is made proportional to its outdegree.

The models studied in this thesis are generalizations of random recursive trees and preferential attachment trees. Hooking networks are constructed recursively, whereby a new graph called a block is attached at each step, instead of a single vertex. Bipolar networks are directed graphs recursively constructed by choosing an arc at random in the network and replacing it with a directed graph. Random recursive metric spaces are similar to hooking networks, but the blocks attached at each step are metric spaces. Finally, a broadcast induced colouring on a random recursive tree or preferential attachment tree is a random 2-colouring of the vertices as red or blue in the following way. The root vertex is coloured red or blue with equal probability, and every other vertex takes the colour of its parent with probability p and the other colour with probability 1-p.

In Paper I, we prove normal limit laws for the degree distributions of hooking networks and bipolar networks. Paper II provides a normal limit law, under certain conditions, for the insertion depth in hooking networks; the distance from the initial starting block to the newly added block. In Paper III, a similar normal limit law is proved for the insertion depth in random recursive metric spaces. Broadcast induced colourings in random recursive trees and preferential attachment trees are studied in Paper IV, where we prove limit laws for the number of vertices of each colour, the number of clusters (maximal monochromatic subtrees) of each colour, as well as the number of leaves of each colour and the number of 2-coloured trees appearing in the fringe. We also prove limit laws for the size of the cluster containing the root vertex.

Place, publisher, year, edition, pages
Uppsala: Department of Mathematics, 2022. , p. 31
Series
Uppsala Dissertations in Mathematics, ISSN 1401-2049 ; 125
National Category
Mathematics
Identifiers
URN: urn:nbn:se:uu:diva-474087ISBN: 978-91-506-2951-4 (print)OAI: oai:DiVA.org:uu-474087DiVA, id: diva2:1656796
Public defence
2022-08-29, Häggsalen, Ångströmlaboratoriet, Lägerhyddsvägen 1, Uppsala, 13:15 (English)
Opponent
Supervisors
Available from: 2022-06-09 Created: 2022-05-08 Last updated: 2022-06-09
List of papers
1. Normal limit laws for vertex degrees in randomly grown hooking networks and bipolar networks
Open this publication in new window or tab >>Normal limit laws for vertex degrees in randomly grown hooking networks and bipolar networks
2020 (English)In: The Electronic Journal of Combinatorics, ISSN 1097-1440, E-ISSN 1077-8926, Vol. 27, no 2, article id P2.45Article in journal (Refereed) Published
Abstract [en]

We consider two types of random networks grown in blocks. Hooking networks are grown from a set of graphs as blocks, each with a labelled vertex called a hook. At each step in the growth of the network, a vertex called a latch is chosen from the hooking network and a copy of one of the blocks is attached by fusing its hook with the latch. Bipolar networks are grown from a set of directed graphs as blocks, each with a single source and a single sink. At each step in the growth of the network, an arc is chosen and is replaced with a copy of one of the blocks. Using Polya urns, we prove normal limit laws for the degree distributions of both networks. We extend previous results by allowing for more than one block in the growth of the networks and by studying arbitrarily large degrees.

Place, publisher, year, edition, pages
ELECTRONIC JOURNAL OF COMBINATORICS, 2020
Keywords
Hooking networks, bipolar networks, central limit laws, Polya urns, random trees, preferential attachment
National Category
Computer Sciences Mathematics
Identifiers
urn:nbn:se:uu:diva-418553 (URN)10.37236/9139 (DOI)000539577000001 ()
Funder
Swedish Research CouncilKnut and Alice Wallenberg Foundation
Available from: 2020-09-04 Created: 2020-09-04 Last updated: 2022-05-08Bibliographically approved
2. Depths in hooking networks
Open this publication in new window or tab >>Depths in hooking networks
2022 (English)In: Probability in the engineering and informational sciences (Print), ISSN 0269-9648, E-ISSN 1469-8951, Vol. 36, no 4, p. 941-949Article in journal (Refereed) Published
Abstract [en]

A hooking network is built by stringing together components randomly chosen from a set of building blocks (graphs with hooks). The vertices are endowed with “affinities” which dictate the attachment mechanism. We study the distance from the master hook to a node in the network chosen according to its affinity after many steps of growth. Such a distance is commonly called the depth of the chosen node. We present an exact average result and a rather general central limit theorem for the depth. The affinity model covers a wide range of attachment mechanisms, such as uniform attachment and preferential attachment, among others. Naturally, the limiting normal distribution is parametrized by the structure of the building blocks and their probabilities. We also take the point of view of a visitor uninformed about the affinity mechanism by which the network is built. To explore the network, such a visitor chooses the nodes uniformly at random. We show that the distance distribution under such a uniform choice is similar to the one under random choice according to affinities.

Place, publisher, year, edition, pages
Cambridge University Press, 2022
Keywords
Distance in graph, Limit law, Network, Preferential attachment, Random graph, Small world
National Category
Mathematics
Identifiers
urn:nbn:se:uu:diva-474088 (URN)10.1017/s0269964821000164 (DOI)000779192000001 ()2-s2.0-85105886943 (Scopus ID)
Funder
Knut and Alice Wallenberg FoundationSwedish Research CouncilEuropean Commission
Available from: 2022-05-08 Created: 2022-05-08 Last updated: 2025-07-18Bibliographically approved
3. Depths in random recursive metric spaces
Open this publication in new window or tab >>Depths in random recursive metric spaces
(English)Manuscript (preprint) (Other academic)
National Category
Mathematics
Identifiers
urn:nbn:se:uu:diva-474090 (URN)
Available from: 2022-05-08 Created: 2022-05-08 Last updated: 2022-05-08
4. Broadcasting induced colourings of random recursive trees and preferential attachment trees
Open this publication in new window or tab >>Broadcasting induced colourings of random recursive trees and preferential attachment trees
(English)Manuscript (preprint) (Other academic)
National Category
Mathematics
Identifiers
urn:nbn:se:uu:diva-474089 (URN)
Available from: 2022-05-08 Created: 2022-05-08 Last updated: 2022-05-08

Open Access in DiVA

UUThesis_C-Desmarais-2022(348 kB)440 downloads
File information
File name FULLTEXT01.pdfFile size 348 kBChecksum SHA-512
fa44f3a5add387f71585b27af05c18212665f5aeaafeb2603d52ee1ae249e968ab93a08dcf0be6449435d5d14db234be31c6689967cab0f1b8bb60168bfd84ec
Type fulltextMimetype application/pdf

Authority records

Desmarais, Colin

Search in DiVA

By author/editor
Desmarais, Colin
By organisation
Department of Mathematics
Mathematics

Search outside of DiVA

GoogleGoogle Scholar
Total: 444 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 1257 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf