The goal in networked control of multiagent systems is to derive desirable collective behavior through the design of local control algorithms. The information available to the individual agents, either through sensing or communication, invariably defines the space of admissible control laws. Hence, informational restrictions impose constraints on the achievable performance guarantees. This talk will focus on how agents should utilize available information to optimize the efficiency of the emergent collective behavior. In particular, we will discuss a methodology for optimizing the efficiency guarantees (i.e., price of anarchy) in distributed resource allocation problems through the design of local agent objective functions. We will then demonstrate the implications of this result in both engineering and societal systems. One particular illustration focuses on the design of taxation mechanisms to optimize the price of anarchy in atomic congestion games.  Here, our findings provide the optimal taxation mechanisms and price of anarchy guarantees for such settings.  

Venue: Games, Decisions & Networks Online Seminar Series

Engineers are often tasked with building the physical infrastructure capable of serving the underlying societal demands. Examples include transportation networks, power grids, data centers, and many more. A fundamental challenge associated with these "socio-technical" systems is that their underlying performance is largely impacted by how society chooses to use them, and unfortunately society tends to use such systems in a highly inefficient way. Jason Marden sheds some light on the unique challenges that surface when seeking to design and control such systems.  

Venue: UCSB Summer GRIT Talks (Ground-Breaking Research Innovative Technology)

The goal in networked control of multiagent systems is to derive desirable collective behavior through the design of local control algorithms. The information available to the individual agents, either through sensing or communication, invariably defines the space of admissible control laws. Hence, informational restrictions impose constraints on the achievable performance guarantees. The first part of this talk will provide one such constraint with regards to the efficiency of the resulting stable solutions for a class of distributed submodular optimization problems.  Further, we will also discuss how strategic information exchange can help mitigate these degradations. The second part of this talk will focus on how agents should utilize available information to optimize the efficiency of the emergent collective behavior. In particular, we will discuss a methodology for optimizing the efficiency guarantees (i.e., price of anarchy) in distributed resource allocation problems through the design of local agent objective functions. Lastly, we will highlight some unintended consequences associated with these optimal designed agent objective functions – optimizing the performance of the worst-case equilibria (i.e., price of anarchy) often comes at the expense of the best-case equilibria (i.e., price of stability).  

Venue: Department of ECE, University of Washington