Incentives for Network Allocation with Strategic Agents: Unifying the Design Process and Aiming at Fairness

Allocation of scarce resources arises in several problems in the general area of networks, including rate allocation in unicast/multicast services on the Internet, transmission-power allocation in wireless networks, spectrum allocation in cellular networks, as well as power/energy production and distribution in the power grid.
Incentive designs (aka Mechanisms) that implement such allocations in the presence of strategic agents have been studied extensively but they invariably suffer from either or both of the following problems:
a) they are usually presented in a case-by-case fashion, thus obscuring fundamental understanding of these problems;
b) they are focusing on the final equilibrium without investing in the process that will lead to such equilibrium, thus propagating the religious notion of "equilibrium economics" .
In addition, these designs almost universally assume that the overall social utility is the sum of agents' utilities. This assumption is mathematically convenient and seemingly value-neutral, but in reality it reflects a bias towards inequality: one "happy" individual in the midst of "unhappy" ones can still maximize the sum of utilities. For this reason it has been vigorously contested since the early days of political economy.
In this talk we ask and offer some answers to the following two questions:
1) Is it possible to unify the incentive design process for a large class of problems and at the same time endow the designs with properties that enable "learning" , i.e., are consistent with a process towards equilibrium?
2) Is it possible to move beyond the "sum-of-utilities" assumption and design mechanisms with fairness as an explicit design goal?