Keywords: strategic, usage, strategic classification, performative prediction, performative
TL;DR: We investigate the repeated game where users strategically choose between a set of learners in pursuit of a positive classification, and show convergence conditions.
Abstract: Real-world systems often involve some pool of users choosing between a set of services. Extensive prior research has been conducted on the effects of strategic users in single-service settings, with strategic behavior manifesting in the manipulation of observable features to achieve a desired classification; however, this can often be costly or unattainable for users and fails to capture the full behavior of multi-service dynamic systems. We analyze a setting in which strategic users choose among several available services in order to pursue positive classifications, while services seek to minimize loss functions on their observations. We show that naive retraining can lead to oscillation even if all users are observed at different times; however, we show necessary and sufficient conditions to guarantee convergent behavior if this retraining uses memory. We provide results obtained from synthetic and real-world data to empirically validate our theoretical findings.
Submission Number: 4
Loading