Path Weights in Undirected Markov Random Fields

Abstract

This research work has focused on statistical inference for network models. Following the direction given in the original submission, the team developed a methodology for paths and path weights. The main aim was to clarify the role played by the path weight in undirected graphical models providing a clear interpretation to the value they take. This goal was successfully achieved and has lead to 5 articles published in top statistical journals. These results were then extended to be made applicable to the analysis of symmetric structures in brain networks form from functional magnetic resonance imaging and the computation of centrality measures in food networks encoding food consumption patterns. In order to increase the dissemination of the research results, Dr. Roverato wrote a book "Graphical Models for Categorical Data", published by Cambridge University Press. Moreover, the research developed under this grant was presented at 14 conferences and professional meetings. Finally, Dr. Roverator co-authored the chapter of a book published by Chapman and Hall "Handbook of Graphical Models".

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 06, 2022
Accession Number
AD1175137

Entities

People

  • Alberto Roverato

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Czech Republic
  • Data Science
  • Gene Expression
  • Information Science
  • Machine Learning
  • Magnetic Resonance
  • Magnetic Resonance Imaging
  • Mathematics
  • Network Science
  • New Zealand
  • Scientific Research
  • Statistical Analysis
  • Statistical Inference
  • Statistics
  • Teamwork
  • Universities

Readers

  • Neural Network Machine Learning.
  • Neuroscience
  • Technical Research and Report Writing.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks