Robustness of Complex Networks: An Approach Using Machine Learning Methods

• Grant number: 25/11306-0
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: August 01, 2025
• End date: July 31, 2028
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Francisco Aparecido Rodrigues
• Grantee: Loriz Francisco Sallum
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Company: Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC)
• Associated research grant: 20/09835-1 - IARA - Artificial Intelligence in the Remaking of Urban Environments, AP.PCPE

Abstract

This research project proposes a comprehensive and innovative investigation into the robustness and resilience of complex networks, areas of growing importance in the study of complex systems, with applications in transportation, data, and power networks. The objective is to develop theoretical and computational methods to analyze and enhance the robustness and resilience of these networks by exploring advanced prediction techniques, topological analysis, and resilience in networks with different topologies. The research will focus on the use of machine learning, particularly neural networks, to predict the behavior of complex networks under various conditions and identify critical failure points. The project aims to investigate how network structure impacts its ability to remain stable in the face of failures and disturbances, proposing solutions to optimize network performance and resilience. The main innovation of this work lies in the integration of network topology studies with advanced machine learning techniques, especially deep neural networks, to predict critical behaviors and vulnerability points. Models will be developed to not only identify sensitive areas in networks with different characteristics but also propose new strategies to mitigate failures and optimize their functionality. In terms of applications, this project aims to make a significant contribution to complex network science by proposing new approaches to predict and strengthen robustness in infrastructure, data, and social networks, enabling a deeper understanding of their resilience against internal and external shocks. In the context of smart cities, this project seeks to apply advancements in the study of complex network robustness to address fundamental urban challenges. Transportation, energy, and communication networks are central components of smart cities, and their interdependence demands robust and resilient strategies to ensure continuous and efficient operation. Using machine learning-based methods, the project aims to identify critical vulnerabilities in these networks and propose solutions that enhance the sustainability and reliability of urban services.Thus, the project aims not only to advance the theoretical understanding of complex networks but also to provide practical tools that support the development of more functional, resilient, and sustainable smart cities. (AU)