Data Science/Computational Social Science Seminar: Amin Rahimian
Mediating social influence and network effects for public health interventions
Register to attend DS/CSS online events
Abstract
I will cover a number of recent results on intervention designs in social network contexts, with a particular focus on public health applications. Peer influence and neighborhood interactions are powerful tools for amplifying the effect of limited, local interventions across the entire scale of a social network. Public health professionals collect social network data to engage with vulnerable populations for improving health outcomes. In “seeding with costly network information” (Operations Research, 2022), we present a unifying framework for data collection and targeted interventions with theoretical guarantees to trade off the cost of data collection with increasing intervention size. Community relations can be harmed if public health efforts pose excessive privacy risks to individuals or fail to effectively address historical and structural inequalities in the population. I will show the utility of our framework in analyzing privacy and fairness implications of data collection and intervention designs. Subsequently, we provide privacy guarantees for influence maximization algorithms when the social network is unknown, and the inputs are samples of prior influence cascades that are collected at random. Building on our sample complexity bounds for seeding with costly network information, our privacy-preserving algorithms introduce randomization in the collected data or the algorithm output, and can bound each node's privacy loss in deciding whether or not their data should be included in the algorithm input. I will end with a collection of empirical results measuring peer effects in the COVID-19 pandemic, as well as recent attempts in using Facebook’s social connectedness index (SCI) to measure the effect of social networks on the spread of the opioid epidemic in the U.S.
Speaker bio
Amin Rahimian joined Pitt IE as an assistant professor in the fall of 2020. Prior to that, he was a postdoc with joint appointments at MIT Institute for Data, Systems, and Society (IDSS) and MIT Sloan School of Management. He received his PhD in electrical and systems engineering from the University of Pennsylvania, and master’s in statistics from Wharton School. Broadly speaking his works are at the intersection of networks, data, and decision sciences. He borrows tools from applied probability, statistics, algorithms, as well as decision and game theory. Some of his current focus is on the challenges of inference and intervention design in complex, large-scale sociotechnical systems, with applications ranging from online social networks, public health, e-commerce and collective decision/action platforms to modern civilian cyberinfrastructure and future battlefields. He is especially interested in the critical role that information plays in the operation of sociotechnical institutions and its societal implications, including on diversity, fairness, and privacy. At Pitt, he leads the sociotechnical systems research lab and teaches courses on stochastic processes, design of experiments, statistics and a new engineering elective on “Data for Social Good (IE 1171)” that he has developed through the Pitt Year of Data and Society Initiative.