Inferring Network Structure and Flows using Partial Observations

Abstract

The overarching goal is to improve the response, resilience, and recovery of interdependent, multilayer physical networks from WMD threats and attacks. The project develops algorithms inspired by network science, physical model, and machine learning that can cope with incomplete observations to predict network structure, network dynamics, and edge flows. Using both physical domain-specific and data-driven approaches, the developed methods address diverse problems including: the reconstruction of network topology and parameters, the estimation of network flows, the optimal selection and scheduling of multiple types of sensors, and the modeling of interdependencies in multiple networks. Together, these methods will build a rigorous theory for solving the inverse problem of edge and edge flow determination from observational behavioral data from critical infrastructure networks. The proposing team consists of experts in network science, machine learning, and modeling of physical systems, and have collaborated extensively on similar problems in the past. Besides the development of necessary methods and theories, the project will also lead to the training of PhD students and postdocs.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 01, 2019
Source ID
HDTRA11910017

Entities

People

  • Ambuj Singh

Organizations

  • Defense Threat Reduction Agency
  • University of California, Santa Barbara

Tags

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Operations Research
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks