How and Why to Manipulate Your Own Agent: On the Incentives of Users of Learning Agents

TL;DR

This paper models strategic interactions among users of learning agents in online economic settings, revealing that users often have incentives to manipulate their agents, which can significantly alter expected outcomes.

Contribution

It introduces a general framework for analyzing strategic manipulation of learning agents in various game settings, highlighting the incentives and effects of such behaviors.

Findings

Users often benefit from misreporting parameters to their agents.

Manipulation can lead to outcomes different from standard predictions.

Strategic behavior influences the equilibrium and efficiency of the system.

Abstract

The usage of automated learning agents is becoming increasingly prevalent in many online economic applications such as online auctions and automated trading. Motivated by such applications, this paper is dedicated to fundamental modeling and analysis of the strategic situations that the users of automated learning agents are facing. We consider strategic settings where several users engage in a repeated online interaction, assisted by regret-minimizing learning agents that repeatedly play a "game" on their behalf. We propose to view the outcomes of the agents' dynamics as inducing a "meta-game" between the users. Our main focus is on whether users can benefit in this meta-game from "manipulating" their own agents by misreporting their parameters to them. We define a general framework to model and analyze these strategic interactions between users of learning agents for general games and analyze the equilibria induced between the users in three classes of games. We show that, generally, users have incentives to misreport their parameters to their own agents, and that such strategic user behavior can lead to very different outcomes than those anticipated by standard analysis.