Advances in Network Analysis and Optimization
Networks are everywhere, from th e financial systems powering global economies to the social connectio ns that shape communities and the infrastructure that moves people an d goods. This dissertation examines the integration of network scienc e, operations research, and machine learning to develop methodologica l contributions that advance the modeling, optimization, and interpre tation of complex systems. These contributions are presented across f our chapters: • Chapter 2 introduces the Feature-Based Network Cons truction (FBNC) framework for reconstructing networks from partial or aggregate data, enabling exact feature-constrained sampling and prov iding new tools for both analysis and “what-if” scenario explorat ion. • Chapter 3 investigates network connectivity optimization thr ough the lens of Markov chains, proposing an algorithm that directly minimizes mean first passage times while remaining robust to uncertai nty in edge presence. • Chapter 4 presents CoNNect, a connectivity- preserving regularization method that enforces computational efficien cy through sparsity in neural networks without sacrificing expressive ness or performance. • Chapter 5 develops a clustering framework to identify the functional positions of financial institutions within m ulti-layer financial networks, offering regulators interpretable insi ghts into systemic roles such as intermediaries, connectors, and peri pheral actors.
More information on the thesis
DESCRIPTION: Advances in Network Analysis and Optimization Networks ar e everywhere, from the financial systems powering global economies to the social connections that shape communities and the infrastructure that moves people and goods. This dissertation examines the integrat ion of network science, operations research, and machine learning to develop methodological contributions that advance the modeling, optim ization, and interpretation of complex systems. These contributions a re presented across four chapters: • Chapter 2 introduces the Featu re-Based Network Construction (FBNC) framework for reconstructing net works from partial or aggregate data, enabling exact feature-constrai ned sampling and providing new tools for both analysis and “what-if ” scenario exploration. • Chapter 3 investigates network connecti vity optimization through the lens of Markov chains, proposing an alg orithm that directly minimizes mean first passage times while remaini ng robust to uncertainty in edge presence. • Chapter 4 presents CoN Nect, a connectivity-preserving regularization method that enforces c omputational efficiency through sparsity in neural networks without s acrificing expressiveness or performance. • Chapter 5 develops a cl ustering framework to identify the functional positions of financial institutions within multi-layer financial networks, offering regulato rs interpretable insights into systemic roles such as intermediaries, connectors, and peripheral actors.More information on the thesis END:VEVENT END:VCALENDAR