Description
Date depot: 1 janvier 1900
Titre: Analysis and modeling of temporal brain networks
Directeur de thèse:
Fabrizio DE VICO FALLANI (ICM)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
The advance of network science has given rise to a tremendous amount of methods and measures for studying complex networks with static topologies. Nevertheless, the translation and generalization of these results to the case where the network exhibits a non-trivial temporally dynamic topology turned out to be not at all trivial: when switching on the time variable as an explicit element of the representation, we are forced to renounce to fundamental properties of the network, e.g. transitivity of edges. In some fortunate cases, the problem has been overcome by projecting the temporal network structure to a static graph but it is well-known that a projection always entails a non-negative loss of information, which in general is not harmless.
The motivation for the study of temporal networks is crystal-clear: in the first place, every realistic network is in some degree a temporal network; secondly, for most of the networks of real interests the time dimension cannot be neglected and it has dramatic consequences on the processes that establish on it: examples can be drawn from economics, social sciences, ecology, biology, and neuroscience where an ever increasing number of systems suitable for a temporal-network approach can be found. Despite the broad interest in them and some (few) efforts, the study of temporal networks represents a fertile research area.
This project has the ambition to develop new theoretical tools for the description of phenomena on temporal networks, ranging from efficient generative models to the understanding of drivers of the network evolution, and validating these models on real data. Among others, spotlight of this project will be on the brain networks, whose temporality features (neuroplasticity) are at the basis of several key functional abilities e.g. learning, recovering from damage. Given the universality of temporal networks across many disciplines, any theoretical advance in this field is likely to be broadly beneficial, from the understanding of epidemiological characteristics of infectious diseases to the rise and fall of collective behaviors in biological processes.
Doctorant.e: Dichio Vito