Statistical physics of information dynamics




Any physical system can be viewed from the perspective that information is implicitly represented in its state. However, the quantification of this information when it comes to complex networks has remained largely elusive. Networked units are ubiquitous, from interacting proteins in cells to neural connections in human brain or social relationships in virtual platforms, such as Twitter or Facebook. They provide a useful tool to model complex systems whose interacting constituents give rise to emergent behavior which can not be caused by single units separately.
In 2016, we have introduced a framework based on the concept of density matrix and the corresponding von Neumann entropy in the case of networks. In the subsequent years, we have further extended and generalized it by using generative models to explicitly account for the interplay between network structure and some dynamics unfolding on its top, shifting the focus from ensembles of networks to ensembles of events on the top of a network.

On the one hand, this framework allowed us to build a ground for an information theory of complex networks (see for further details). On the other hand the framework can be used to define variational methods to describe the emergence of sparsity in complex networks or the emergence of ubiquitous topological features. Recently, the framework has been used to introduce a suitable renormalization group for complex networks.

Besides enhancing our understanding of complex systems from a statistical physics perspective, applications include machine learning, systems biology and inference of quantum complex network models based on qubit entanglement. To date, we have applied the framework to analyze biomolecular networks, connectomes, fungal networks, social networks and network infrastructures.


Relevant publications

Diversity of information pathways drives sparsity in real-world networks
A. Ghavasieh, M. De Domenico, Nature Physics (2024)
Abstract »   Read »   BibTeX
 
   

Emergent information dynamics in many-body interconnected systems
W. Merbis, M. De Domenico, Phys. Rev. E 108, 014312 (2023)
Abstract »   Read »   BibTeX
 
   

Generalized network density matrices for analysis of multiscale functional diversity
A. Ghavasieh, M. De Domenico, Phys. Rev. E 107, 044304 (2023)
Abstract »   Read »   BibTeX
 
   

Dismantling the information flow in complex interconnected systems
A. Ghavasieh, G. Bertagnolli, M. De Domenico, Phys. Rev. Research 5, 013084 (2023)
Abstract »   Read »   BibTeX
   

Maximum entropy network states for coalescence processes
A. Ghavasieh, M. De Domenico,  (2022)
Abstract »   Read »  
  
   

Multiscale Information Propagation in Emergent Functional Networks
A. Ghavasieh, M. De Domenico, Entropy 23, 1369 (2021)
Abstract »   Read »   BibTeX
  
   

Persistence of information flow: a multiscale characterization of human brain
B. Benigni, A. Ghavasieh, A. Corso, V. d'Andrea, M. De Domenico, Network Neuroscience 5, 831 (2021)
Abstract »   Read »   BibTeX
 
   

Unraveling the effects of multiscale network entanglement on empirical systems
A. Ghavasieh, M. Stella, J. Biamonte, M. De Domenico, Communications Physics 4, 129 (2021)
Abstract »   Read »   BibTeX
 
   

Multiscale statistical physics of the pan-viral interactome unravels the systemic nature of SARS-CoV-2 infections
A. Ghavasieh, S. Bontorin, O. Artime, N. Verstraete, M. De Domenico, Communications Physics 4, 83 (2021)
Abstract »   Read »   BibTeX
 
   

Enhancing transport properties in interconnected systems without altering their structure
A. Ghavasieh, M. De Domenico, Phys. Rev. Research 2, 013155 (2020)
Abstract »   Read »   BibTeX
   

Spectral entropies as information-theoretic tools for complex network comparison
M. De Domenico, J. Biamonte, Phys. Rev. X 6, 041062 (2016)
Abstract »   Read »   BibTeX
 
   


For a more comprehensive list of papers about this topic, including the most recent theoretical developments and applications from cells to societies, click here.