Organizer(s):

Giuseppe Ferro, Princeton University

Lazaros Gallos, DIMACS

Simon Levin, Princeton University

David Pennock, DIMACS

Fred Roberts, DIMACS

Will Tracy, Santa Fe Institute

Emma Zajdela, Princeton University

Description

Understanding the dynamics of how physical and social processes spread on complex networks is a multifaceted question with theoretical and empirical implications. From an application perspective, this question is relevant in many fields, from the spread of disease, opinions, and innovation to environmental issues. For example, identifying methods to facilitate spreading of virtuous behaviors in social networks effectively and rapidly could help mitigate the effects of climate change. Conversely, understanding how to slow spreading on networks could help curtail the spread of misinformation or disease. The recent supply chain crisis during the Covid-19 pandemic underscored the importance of designing robust networks that are resistant to systemic risk. These networks are often not static, but the nodes and links evolve concurrently with the spreading processes. Incentives play a role too: rewards and penalties can steer people toward positive behaviors and slow the spread of bad outcomes. Furthermore, these processes may occur over multiple temporal, spatial or other scales (e.g. individual versus institutional).

From a theoretical perspective, several questions of interest include: Where in a network might one seed a process to have it spread most rapidly? Similarly, how can we identify points or links whose suppression would slow spread? How can we design robust networks and control to resist systemic risk? If the goal is to control a spreading process (such as contagion) over a graph, what are the relevant considerations for having centralized versus distributed control over the spreading process? How are the tradeoffs each approach modulated by different graph topologies? How can networks be designed to incentivize good behavior and discourage bad behavior? How are these processes affected in multilayer or multilevel networks with varying temporal or physical scales? 

Studying these questions requires an interdisciplinary perspective, drawing on insights from various fields including (but not limited to) network science, graph theory, control theory, economics, and dynamical systems, as well expertise within the domain of applications.

This event is co-sponsored by the Center for BioComplexity, High Meadows Environmental Institute, Princeton University and The Santa Fe Institute.

Program

Monday, October 21, 2024

Tuesday, October 22, 2024

Wednesday, October 23, 2024

Call For Participation

The workshop will be a gathering of around 30 invited participants to allow for meaningful conversations and interaction. The structure of the workshop will include keynote lectures, technical talks focused on particular methodologies or computational techniques, flash talks, breakout groups throughout the workshop, and ample informal time for discussion and possible collaborations to form.

Sponsored by Santa Fe Institute, Center for BioComplexity, High Meadows environmental institute, Princeton University, in association with the SF on Mechanisms & Algorithms to Augment Human Decision Making.