Fiscko and Blanco awarded NSF Graduate Research Fellowships
Two Ph.D. students from Carnegie Mellon University’s Department of Electrical and Computer Engineering have received 2019 National Science Foundation (NSF) Graduate Research Fellowships. For the next 3 years, Carmel Fiscko and Mark Blanco will each receive an annual awards of $12,000 in tuition and a $34,000 stipend to continue their innovative research.
Carmel Fiscko’s work focuses on understanding how agents within a network make decisions and command influence. Her early findings have demonstrated how global influencers can understand the long-term effects of their attempts to influence other users within a network.
As an NSF graduate research fellow, Fiscko will work on measuring levels of inter-agent influence from the perspective of an outside observer, and understanding how ideas and actions propagate throughout a network. Future work may develop tools to predict how agents react to new stimuli or tackle other adversarial agents.
Her work has wide implications in both social networks, and networks in general, such as security/usage of computer networks, traffic networks, autonomous robotics, IoT, and economics.
Mark Blanco works in analytical performance modeling for high performance computing (HPC). His focus is on graph analytic algorithms—programs that operate on a network representation of objects or individuals connected by relationships.
Blanco’s research addresses the challenge of how to implement graph analytics for increasingly complex data on a growing variety of computing platforms, including datacenters, desktops, mobile phones, and the Internet of Things. Graph analytics are an important tool in areas such as IT security, chemistry, neuroscience, and in data-powered services many of us use daily, but the difficulty in scaling graph algorithms to meet today’s massive data demands has limited the deployment of graph analytics.
His goal is to develop models that capture how graph algorithms execute on and interact with the intricate internals of current and future platform architectures. Such models would guide programmers in how to best implement graph analytics and organize the data used by them, and could help direct the design of future architectures and hardware.
Mark Blanco is advised by Assistant Research Professor Tze Meng Low. Carmel Fiscko is co-advised by Professor Bruno Sinopoli and Associate Professor Soummya Kar.