Click here to start your application. Apply now

Network Science Institute London

In 2022, Network Science Institute (NetSI) established a new institute hub site in London, UK.

The London hub will serve as a dynamic platform fostering the seamless exchange of knowledge and expertise between the United States and Europe. Its purpose extends beyond traditional collaboration; it serves as a key centre where students and scholars can effortlessly connect and collaborate across geographical boundaries.

In addition to bringing fresh perspectives and innovative methodologies into established research domains like (i) Network Epidemiology and Forecasting and (ii) Fundamental Network Science, the research conducted in London is primed to catalyse growth in pivotal areas such as: (iii) Urban dynamics and Computational Social Science, (iv) Networks and AI, and (v) Network neuroscience.

Main Research Areas

Our research in urban dynamics and computational social science explores the intricate relationships between cities and its citizens through advanced data analysis and modelling. In an increasingly interconnected world, individuals generate vast amounts of digital data through their activities, interactions, and movements. We leverage the digital fingerprint of citizens to study human behaviour and social interactions. We adopt a multidisciplinary approach that combines elements of complex science, network theory, and machine learning to better comprehend the complex ecosystem shaped by people’s actions and interactions in cities and online.

Our research focuses on understanding the role of genetic, structural, and functional connectivity in performance, regulation, and disease. By mapping relations across multiple imaging modalities and genotypic features of brains across species, we aim to build data-based and theoretically-driven models of higher cognitive functions to improve modelling of loss and recovery of function.

Our research focuses on exploring how artificial intelligence systems make sense of the world based on sparse observations, and how they learn structures and symmetries. Networks and related topological approaches provide a rigorous mathematical framework to study the symmetries and transformations inherent in information processing within complex systems. By leveraging these concepts, we aim to unravel the intricate relationship between the networked structure of functional spaces within neural networks with their performances, limits, and robustness.

Despite outstanding theoretical advances over the last two decades, many problems still lie open both in theoretical areas, as well as in existing fields of application of network science. This is compounded by the recent emergence of generalised network structures, such as multilayered, temporal, and higher-order networks. In addition, empirical observations and intuitions often unravel surprising results whose theoretical/rigorous underpinning are not evident yet. There is therefore ample scope for principled and rigorous reasoning. Our research aims to close this gaps between theory and applications.

Much of the infrastructure around us, both real and virtual, such as telecoms, shipping and power networks are amenable to be modelled as networks or be analysed using tools originating in network science. By leveraging such tools, our research unravels patterns and understands interdependencies between network properties and system performance to aid better design and efficient interventions.