Data-driven Dynamic Intervention Design in Network Games

TL;DR

This paper develops a data-driven intervention method for network games that guides agents towards desired actions without prior knowledge of utilities or network parameters, ensuring convergence and considering intervention budgets.

Contribution

It introduces a novel data-driven intervention mechanism that operates without prior utility or network information, with proven convergence guarantees and practical effectiveness.

Findings

Mechanism successfully guides agents to target actions.

Convergence guarantees are analytically established.

Numerical case study confirms effectiveness.

Abstract

Targeted interventions in games present a challenging problem due to the asymmetric information available to the regulator and the agents. This note addresses the problem of steering the actions of self-interested agents in quadratic network games towards a target action profile. A common starting point in the literature assumes prior knowledge of utility functions and/or network parameters. The goal of the results presented here is to remove this assumption and address scenarios where such a priori knowledge is unavailable. To this end, we design a data-driven dynamic intervention mechanism that relies solely on historical observations of agent actions and interventions. Additionally, we modify this mechanism to limit the amount of interventions, thereby considering budget constraints. Analytical convergence guarantees are provided for both mechanisms, and a numerical case study further demonstrates their effectiveness.