서브메뉴
검색
Compatibility, Predictions and Optimization in Large-Scale Online Decision Making
Compatibility, Predictions and Optimization in Large-Scale Online Decision Making
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105554
- ISBN
- 9798263398033
- DDC
- 330
- 저자명
- Rutten, Daan.
- 서명/저자
- Compatibility, Predictions and Optimization in Large-Scale Online Decision Making
- 발행사항
- [Sl] : Georgia Institute of Technology, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 290 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Mukherjee, Debankur.
- 학위논문주기
- Thesis (Ph.D.)--Georgia Institute of Technology, 2024.
- 초록/해제
- 요약The last few years have seen a shift in server architecture towards large-scale, dedicateddata centers. The massive scale of these datacenters has highlighted important, unsolvedproblems in the space of load balancing, speed scaling and related areas in the optimizationof such large-scale service systems. Many of the mathematical problems abstracted fromthese challenges form fundamental problems in the field of online decision making andrequire a combination of tools from probability, stochastic processes and online algorithmstogether with completely new insights to be solvable. This thesis broadly addresses threeof these problems, as detailed below.In Chapter 2, we consider a large-scale, parallel-server system, where tasks of a particular type can only be routed to a small subset of servers. The task-server constraints arerepresented by a bipartite graph where vertices represent task types and servers, respectively. The analysis of these systems has historically relied heavily on mean-field analysis.A pivotal assumption for this framework is that servers are exchangeable. However, due tothe lack of exchangeability in the constrained system, mean-field techniques fundamentallybreak down. In this chapter, we develop a novel coupling-based approach to establish themean-field approximation for a large class of sparse graphs, including spatial graphs. Themethod extends the scope of mean-field analysis far beyond the classical full-flexibilitysetup.In Chapter 3, we consider a large-scale, parallel-server system with an unknown arrivalrate, where each server is able to adjust its processing speed. The objective is to minimize the system cost, which consists of a power cost to maintain the servers' processingspeed and a quality of service cost depending on the tasks' sojourn times. We draw onideas from stochastic approximation to design a novel speed scaling algorithm and provethat the server's processing speeds converge to the globally asymptotically optimum value.Curiously, the algorithm is fully distributed and does not require any communication between servers. Apart from the algorithm design, a key contribution of our approach lies indemonstrating how concepts from the stochastic approximation literature can be leveragedto effectively tackle learning problems in large-scale, distributed systems.In Chapter 4, we consider learning-augmented algorithms, where the decision makerhas access to a black-box oracle, such as a machine learning model, that provides untrustedand potentially inaccurate predictions of future inputs. The goal of the decision maker isto exploit the predictions if they are accurate, while guaranteeing performance that is notmuch worse than the hindsight optimal sequence of decisions, even when predictions areinaccurate. We consider two applications: capacity scaling and smoothed online optimization. For both applications, we design a novel algorithm and prove a competitive ratioguarantee as a function of the predictions' accuracy. Interestingly, we identify a fundamental trade-off between the worst-case (predictions are inaccurate) and best-case (predictionsare accurate). In fact, we prove that this trade-off is necessary for any algorithm.
- 일반주제명
- Sparsity
- 일반주제명
- Graphs
- 일반주제명
- Neural networks
- 일반주제명
- Decision making
- 일반주제명
- Stochastic models
- 일반주제명
- Processing speed
- 일반주제명
- Mathematics
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2024 us c eng d■001000017360604
■00520260202105554
■006m o d
■007cr#unu||||||||
■020 ▼a9798263398033
■035 ▼a(MiAaPQ)AAI32315845
■035 ▼a(MiAaPQ)GeorgiaTech75185
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a330
■1001 ▼aRutten, Daan.
■24510▼aCompatibility, Predictions and Optimization in Large-Scale Online Decision Making
■260 ▼a[Sl]▼bGeorgia Institute of Technology▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a290 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Mukherjee, Debankur.
■5021 ▼aThesis (Ph.D.)--Georgia Institute of Technology, 2024.
■520 ▼aThe last few years have seen a shift in server architecture towards large-scale, dedicateddata centers. The massive scale of these datacenters has highlighted important, unsolvedproblems in the space of load balancing, speed scaling and related areas in the optimizationof such large-scale service systems. Many of the mathematical problems abstracted fromthese challenges form fundamental problems in the field of online decision making andrequire a combination of tools from probability, stochastic processes and online algorithmstogether with completely new insights to be solvable. This thesis broadly addresses threeof these problems, as detailed below.In Chapter 2, we consider a large-scale, parallel-server system, where tasks of a particular type can only be routed to a small subset of servers. The task-server constraints arerepresented by a bipartite graph where vertices represent task types and servers, respectively. The analysis of these systems has historically relied heavily on mean-field analysis.A pivotal assumption for this framework is that servers are exchangeable. However, due tothe lack of exchangeability in the constrained system, mean-field techniques fundamentallybreak down. In this chapter, we develop a novel coupling-based approach to establish themean-field approximation for a large class of sparse graphs, including spatial graphs. Themethod extends the scope of mean-field analysis far beyond the classical full-flexibilitysetup.In Chapter 3, we consider a large-scale, parallel-server system with an unknown arrivalrate, where each server is able to adjust its processing speed. The objective is to minimize the system cost, which consists of a power cost to maintain the servers' processingspeed and a quality of service cost depending on the tasks' sojourn times. We draw onideas from stochastic approximation to design a novel speed scaling algorithm and provethat the server's processing speeds converge to the globally asymptotically optimum value.Curiously, the algorithm is fully distributed and does not require any communication between servers. Apart from the algorithm design, a key contribution of our approach lies indemonstrating how concepts from the stochastic approximation literature can be leveragedto effectively tackle learning problems in large-scale, distributed systems.In Chapter 4, we consider learning-augmented algorithms, where the decision makerhas access to a black-box oracle, such as a machine learning model, that provides untrustedand potentially inaccurate predictions of future inputs. The goal of the decision maker isto exploit the predictions if they are accurate, while guaranteeing performance that is notmuch worse than the hindsight optimal sequence of decisions, even when predictions areinaccurate. We consider two applications: capacity scaling and smoothed online optimization. For both applications, we design a novel algorithm and prove a competitive ratioguarantee as a function of the predictions' accuracy. Interestingly, we identify a fundamental trade-off between the worst-case (predictions are inaccurate) and best-case (predictionsare accurate). In fact, we prove that this trade-off is necessary for any algorithm.
■590 ▼aSchool code: 0078.
■650 4▼aSparsity
■650 4▼aGraphs
■650 4▼aNeural networks
■650 4▼aDecision making
■650 4▼aStochastic models
■650 4▼aOrdinary differential equations
■650 4▼aProcessing speed
■650 4▼aMathematics
■690 ▼a0800
■690 ▼a0405
■71020▼aGeorgia Institute of Technology.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0078
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360604▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


