Strategic Representation

TL;DR

This paper develops algorithms for decision-making systems that are robust against strategic manipulation of representations, balancing truthful information disclosure with resistance to user deception.

Contribution

It formalizes the problem of strategic representation learning and introduces algorithms that minimize error under strategic manipulation, with theoretical analysis of the trade-offs involved.

Findings

Proposed a learning algorithm resilient to strategic manipulation.

Analyzed the trade-off between learning effort and manipulation susceptibility.

Provided theoretical guarantees for robust decision-making.

Abstract

Humans have come to rely on machines for reducing excessive information to manageable representations. But this reliance can be abused -- strategic machines might craft representations that manipulate their users. How can a user make good choices based on strategic representations? We formalize this as a learning problem, and pursue algorithms for decision-making that are robust to manipulation. In our main setting of interest, the system represents attributes of an item to the user, who then decides whether or not to consume. We model this interaction through the lens of strategic classification (Hardt et al. 2016), reversed: the user, who learns, plays first; and the system, which responds, plays second. The system must respond with representations that reveal `nothing but the truth' but need not reveal the entire truth. Thus, the user faces the problem of learning set functions under strategic subset selection, which presents distinct algorithmic and statistical challenges. Our main result is a learning algorithm that minimizes error despite strategic representations, and our theoretical analysis sheds light on the trade-off between learning effort and susceptibility to manipulation.