Advanced search
Start date

Artificial neural networks and complex networks: an integrative study of topological properties and pattern recognition

Grant number: 19/07811-0
Support type:Scholarships in Brazil - Doctorate
Effective date (Start): January 01, 2020
Effective date (End): July 31, 2022
Field of knowledge:Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
Principal Investigator:Odemir Martinez Bruno
Grantee:Leonardo Felipe dos Santos Scabini
Home Institution: Instituto de Física de São Carlos (IFSC). Universidade de São Paulo (USP). São Carlos , SP, Brazil


Artificial Neural Networks (ANN) and complex networks are gaining attention nowadays due to advances in artificial intelligence, computer hardware, and big data phenomena. Recent deep ANN architectures have been achieving outstanding performance on complex problems, however, due to the lack of knowledge of its internal functioning, these techniques are used as a "black box" approach. In this sense new studies are needed regarding the functioning of these networks, contributing to the development of more robust and interpretable models. On the other hand, the analysis of large complex networks is also a mathematical and computational challenge, and it is essential in the understanding of several real systems that govern modern society. In this context, the proposal for this research is an integrative study between both areas. ANN are naturally represented by graphs, and their entities (neurons) actively interact between themselves. Therefore, the topological properties of ANN will be analyzed through complex network techniques to establish correlations between its structural organization and functioning. The other line of research in the proposal is to use and modify recent convolutional ANN architectures for pattern recognition in complex networks, such as the characterization of different topological classes. Multiscale and dynamic analysis techniques in complex networks will be approached for the construction of smaller representations, thus obtaining matrices that can be used as input of convolutional ANN. In short, the goal is to raise new knowledge in both areas, as well as the development of new models and techniques combining ANN and complex networks for pattern recognition. Furthermore, the proposals will be applied in different real problems through national and international research partners of areas such as Biology, Medicine, and Agriculture. (AU)