Graph Design via Convex Optimization: Online and Distributed Perspectives
Network and graph have long been natural abstraction of relations in a variety of applications, e.g. transportation, power system, social network, communication, electrical circuit, etc. As a large number of computation and optimization problems are naturally defined on graphs, graph structures not...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Dissertation |
| Sprache: | Englisch |
| Veröffentlicht: |
ProQuest Dissertations & Theses
01.01.2017
|
| Schlagworte: | |
| ISBN: | 9780355119206, 035511920X |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | Network and graph have long been natural abstraction of relations in a variety of applications, e.g. transportation, power system, social network, communication, electrical circuit, etc. As a large number of computation and optimization problems are naturally defined on graphs, graph structures not only enable important properties of these problems, but also leads to highly efficient distributed and online algorithms. For example, graph separability enables the parallelism for computation and operation as well as limits the size of local problems. More interestingly, graphs can be defined and constructed in order to take best advantage of those problem properties. This dissertation focuses on graph structure and design in newly proposed optimization problems, which establish a bridge between graph properties and optimization problem properties. We first study a new optimization problem called Geodesic Distance Maximization Problem (GDMP). Given a graph with fixed edge weights, finding the shortest path, also known as the geodesic, between two nodes is a well-studied network flow problem. We introduce the Geodesic Distance Maximization Problem (GDMP): the problem of finding the edge weights that maximize the length of the geodesic subject to convex constraints on the weights. We show that GDMP is a convex optimization problem for a wide class of flow costs, and provide a physical interpretation using the dual. We present applications of the GDMP in various fields, including optical lens design, network interdiction, and resource allocation in the control of forest fires. We develop an Alternating Direction Method of Multipliers (ADMM) by exploiting specific problem structures to solve large-scale GDMP, and demonstrate its effectiveness in numerical examples. We then turn our attention to distributed optimization on graph with only local communication. Distributed optimization arises in a variety of applications, e.g. distributed tracking and localization, estimation problems in sensor networks, multi-agent coordination. Distributed optimization aims to optimize a global objective function formed by summation of coupled local functions over a graph via only local communication and computation. We developed a weighted proximal ADMM for distributed optimization using graph structure. This fully distributed, single-loop algorithm allows simultaneous updates and can be viewed as a generalization of existing algorithms. More importantly, we achieve faster convergence by jointly designing graph weights and algorithm parameters. Finally, we propose a new problem on networks called Online Network Formation Problem: starting with a base graph and a set of candidate edges, at each round of the game, player one first chooses a candidate edge and reveals it to player two, then player two decides whether to accept it; player two can only accept limited number of edges and make online decisions with the goal to achieve the best properties of the synthesized network. The network properties considered include the number of spanning trees, algebraic connectivity and total effective resistance. These network formation games arise in a variety of cooperative multiagent systems. We propose a primal-dual algorithm framework for the general online network formation game, and analyze the algorithm performance by the competitive ratio and regret. |
|---|---|
| AbstractList | Network and graph have long been natural abstraction of relations in a variety of applications, e.g. transportation, power system, social network, communication, electrical circuit, etc. As a large number of computation and optimization problems are naturally defined on graphs, graph structures not only enable important properties of these problems, but also leads to highly efficient distributed and online algorithms. For example, graph separability enables the parallelism for computation and operation as well as limits the size of local problems. More interestingly, graphs can be defined and constructed in order to take best advantage of those problem properties. This dissertation focuses on graph structure and design in newly proposed optimization problems, which establish a bridge between graph properties and optimization problem properties. We first study a new optimization problem called Geodesic Distance Maximization Problem (GDMP). Given a graph with fixed edge weights, finding the shortest path, also known as the geodesic, between two nodes is a well-studied network flow problem. We introduce the Geodesic Distance Maximization Problem (GDMP): the problem of finding the edge weights that maximize the length of the geodesic subject to convex constraints on the weights. We show that GDMP is a convex optimization problem for a wide class of flow costs, and provide a physical interpretation using the dual. We present applications of the GDMP in various fields, including optical lens design, network interdiction, and resource allocation in the control of forest fires. We develop an Alternating Direction Method of Multipliers (ADMM) by exploiting specific problem structures to solve large-scale GDMP, and demonstrate its effectiveness in numerical examples. We then turn our attention to distributed optimization on graph with only local communication. Distributed optimization arises in a variety of applications, e.g. distributed tracking and localization, estimation problems in sensor networks, multi-agent coordination. Distributed optimization aims to optimize a global objective function formed by summation of coupled local functions over a graph via only local communication and computation. We developed a weighted proximal ADMM for distributed optimization using graph structure. This fully distributed, single-loop algorithm allows simultaneous updates and can be viewed as a generalization of existing algorithms. More importantly, we achieve faster convergence by jointly designing graph weights and algorithm parameters. Finally, we propose a new problem on networks called Online Network Formation Problem: starting with a base graph and a set of candidate edges, at each round of the game, player one first chooses a candidate edge and reveals it to player two, then player two decides whether to accept it; player two can only accept limited number of edges and make online decisions with the goal to achieve the best properties of the synthesized network. The network properties considered include the number of spanning trees, algebraic connectivity and total effective resistance. These network formation games arise in a variety of cooperative multiagent systems. We propose a primal-dual algorithm framework for the general online network formation game, and analyze the algorithm performance by the competitive ratio and regret. |
| Author | Meng, De |
| Author_xml | – sequence: 1 givenname: De surname: Meng fullname: Meng, De |
| BookMark | eNotjcFKxDAUAAMqqGv_IeB5IS95aRNv0tV1YaEe9Lwk7atG1rQ2aRG_3gW9zNxmrtl5HCKdscJWRiitAawU5SUrUgpeCGGVEiiv2G47ufGdbyiFt8iX4Hg9xIW-eTPm8Bl-XA5DvONNPIZI3MWOb0LKU_Bzpo4_05RGanNYKN2wi94dExX_XrHXx4eX-mm9b7a7-n6_9iABT6yoN9Ki9lYAuqoryZDUXvjWVmA6TdhLT33VAggrTIkee9-hcsY4A2rFbv-64zR8zZTy4WOYp3haHsAioNZWo_oFNNxK0g |
| ContentType | Dissertation |
| Copyright | Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works. |
| Copyright_xml | – notice: Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works. |
| DBID | 053 0BH 0PY CBPLH EU9 G20 M8- PHGZT PKEHL PQEST PQQKQ PQUKI |
| DatabaseName | Dissertations & Theses Europe Full Text: Science & Technology ProQuest Dissertations and Theses Professional Dissertations & Theses @ University of Washington WCLP ProQuest Dissertations & Theses Global: The Sciences and Engineering Collection ProQuest Dissertations & Theses A&I ProQuest Dissertations & Theses Global ProQuest Dissertations and Theses A&I: The Sciences and Engineering Collection ProQuest One Academic (New) ProQuest One Academic Middle East (New) ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic (retired) ProQuest One Academic UKI Edition |
| DatabaseTitle | Dissertations & Theses Europe Full Text: Science & Technology ProQuest One Academic Middle East (New) ProQuest One Academic UKI Edition ProQuest One Academic Eastern Edition Dissertations & Theses @ University of Washington WCLP ProQuest Dissertations & Theses Global: The Sciences and Engineering Collection ProQuest Dissertations and Theses Professional ProQuest One Academic ProQuest Dissertations & Theses A&I ProQuest One Academic (New) ProQuest Dissertations and Theses A&I: The Sciences and Engineering Collection ProQuest Dissertations & Theses Global |
| DatabaseTitleList | Dissertations & Theses Europe Full Text: Science & Technology |
| Database_xml | – sequence: 1 dbid: G20 name: ProQuest Dissertations & Theses Global url: https://www.proquest.com/pqdtglobal1 sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Mathematics |
| ExternalDocumentID | 4323879427 |
| Genre | Dissertation/Thesis |
| GroupedDBID | 053 0BH 0PY ABAPV CBPLH EU9 G20 M8- PHGZT PKEHL PQEST PQQKQ PQUKI |
| ID | FETCH-LOGICAL-b1214-b17ef82945b9014a7d6e8e25b0bc9718d5e4f2bef7c11090864b4fbd43a88a813 |
| IEDL.DBID | G20 |
| ISBN | 9780355119206 035511920X |
| IngestDate | Mon Jun 30 04:22:59 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-b1214-b17ef82945b9014a7d6e8e25b0bc9718d5e4f2bef7c11090864b4fbd43a88a813 |
| Notes | SourceType-Dissertations & Theses-1 ObjectType-Dissertation/Thesis-1 content type line 12 |
| PQID | 1941455954 |
| PQPubID | 18750 |
| ParticipantIDs | proquest_journals_1941455954 |
| PublicationCentury | 2000 |
| PublicationDate | 20170101 |
| PublicationDateYYYYMMDD | 2017-01-01 |
| PublicationDate_xml | – month: 01 year: 2017 text: 20170101 day: 01 |
| PublicationDecade | 2010 |
| PublicationYear | 2017 |
| Publisher | ProQuest Dissertations & Theses |
| Publisher_xml | – name: ProQuest Dissertations & Theses |
| SSID | ssib000933042 ssib034632569 |
| Score | 1.7456874 |
| Snippet | Network and graph have long been natural abstraction of relations in a variety of applications, e.g. transportation, power system, social network,... |
| SourceID | proquest |
| SourceType | Aggregation Database |
| SubjectTerms | Electrical engineering Mathematics |
| Title | Graph Design via Convex Optimization: Online and Distributed Perspectives |
| URI | https://www.proquest.com/docview/1941455954 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV27TsMwFL2CwgAMvMWjIA-sEa1jJzYLQ0uBgdIBpG6Vn1IHUmhKxefj6ya0EhILSyQrVmTZyj2-r3MAroLHYExqVEJTlyaMK5MIpkXC84B9Ovem5XQUm8j7fTEcykEVcCurssraJkZDbScGY-TXwdlGTm3J2e37R4KqUZhdrSQ01mEDu2tjs-_q9WfhrdfjNMwICC9j0pZjAo22ahKeepz9sskRaHq7_13iHux0VzLs-7DmigPYfvohZy0P4fEeSapJN9ZukPlYkQ6Wnn-R52A-3qq-zBuyICElqrCki-y6KIzlLBksuzPLI3jt3b10HpJKUSHRbdpm4Zk7L6hkXGP6VOU2c8JRrlvayIBSljvmqXY-N8hEGtwdppnXlqVKCCXa6TE0iknhToAIZTLPteWGKqa4lN6w8A0kIg-TvT-FZr1Jo-q3KEfLHTr7-_U5bFHEzxjraEJjNv10F7Bp5rNxOb2Mp_wNypGwBA |
| linkProvider | ProQuest |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1JTwIxFH5BNHE5uMcFtQc9ToROy3RMjAcQISxywIQbtp024SAog6h_yt9o38AIiYk3D14maWbSZOZ13v6-D-DcRQxa-1p61De-x7jUnmBKeDxwtk8FVueNSsgmglZLdLthOwOf6SwMtlWmOjFR1NFQY4780gXbiKkdcnbz_OIhaxRWV1MKjemxqJuPNxeyxde1spPvBaWV206p6s1YBTxVoAXmroGxgoaMKywhyiAqGmEoV3mlQ6epI26YpcrYQCMap3P5mWJWRcyXQkhR8N2-S7DMEOoOh4sX3a1pdiBd-6zoO48iTIrEHAt2NJ-C_qTr4g8bkBi2yuZ_-yRbsFFe6CDYhowZ7MB68xt8Nt6F2h2CcJNy0ptCJn1JStha_07unXp8ms2dXpEpyCqRg4iUET0Yib9MRNrz6dN4Dx7-5EX2ITsYDswBECF10XIVcU0lkzwMrWZuDwRadw9bewi5VCi92W8f9-YSOfr99hmsVjvNRq9Ra9WPYY2ir5DkdXKQHY9ezQms6Mm4H49OkxNG4PGv5fcF868L8A |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adissertation&rft.genre=dissertation&rft.title=Graph+Design+via+Convex+Optimization%3A+Online+and+Distributed+Perspectives&rft.DBID=053%3B0BH%3B0PY%3BCBPLH%3BEU9%3BG20%3BM8-%3BPHGZT%3BPKEHL%3BPQEST%3BPQQKQ%3BPQUKI&rft.PQPubID=18750&rft.au=Meng%2C+De&rft.date=2017-01-01&rft.pub=ProQuest+Dissertations+%26+Theses&rft.isbn=9780355119206&rft.externalDBID=HAS_PDF_LINK&rft.externalDocID=4323879427 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9780355119206/lc.gif&client=summon&freeimage=true |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9780355119206/mc.gif&client=summon&freeimage=true |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9780355119206/sc.gif&client=summon&freeimage=true |