Automated distributed computation and decision-making are common tasks in large-scale autonomous systems. This talk will explore how system designers can endow autonomous agents with local decision-making rules in order to jointly maximize an objective function. Specifically, we will focus on a set of distributed greedy algorithms that address a submodular objective function. Our results show precisely how optimality degrades in the presence of information sharing constraints which arise from computational, time, bandwidth, or security limitations.

David Grimsman is an Assistant Professor in the Computer Science Department at Brigham Young University (BYU). He received a PhD from the Electrical and Computer Engineering Department at UC Santa Barbara in 2021, being advised by Professor Jason Marden and Professor João Hespanha. Prior to that, he received MS in Computer Science and a BS in Electrical and Computer Engineering, both from BYU. His research interests are centered on multiagent systems, game theory, robustness, security, and sports analytics. He is currently the Director of IDeA Labs at BYU, an interdisciplinary research group focus on applying notions of modeling and robustness to several application domains.