Price of Transparency in Strategic Machine Learning

TL;DR

This paper investigates how transparency in machine learning algorithms affects their performance when users are strategic, introducing the concept of the 'price of transparency' through game-theoretic modeling.

Contribution

It models the strategic classification problem as a nonzero-sum game and quantifies the cost of transparency in terms of the price of transparency metric.

Findings

Quantifies the price of transparency in strategic classification.

Models the interaction as a Stackelberg game.

Provides insights into the trade-offs between transparency and performance.

Abstract

Based on the observation that the transparency of an algorithm comes with a cost for the algorithm designer when the users (data providers) are strategic, this paper studies the impact of strategic intent of the users on the design and performance of transparent ML algorithms. We quantitatively study the {\bf price of transparency} in the context of strategic classification algorithms, by modeling the problem as a nonzero-sum game between the users and the algorithm designer. The cost of having a transparent algorithm is measured by a quantity, named here as price of transparency which is the ratio of the designer cost at the Stackelberg equilibrium, when the algorithm is transparent (which allows users to be strategic) to that of the setting where the algorithm is not transparent.