Incentive-Aware Machine Learning; Robustness, Fairness, Improvement & Causality

TL;DR

This paper reviews incentive-aware machine learning, focusing on robustness, fairness, and causality, and introduces a unified framework to address strategic behavior in decision-making systems.

Contribution

It provides a comprehensive categorization and a unified framework for incentive-aware ML, integrating robustness, fairness, and causal perspectives with practical and theoretical insights.

Findings

Models can be made resilient to gaming strategies

Fairness considerations are crucial in strategic settings

Causal approaches help distinguish genuine improvements

Abstract

The article explores the emerging domain of incentive-aware machine learning (ML), which focuses on algorithmic decision-making in contexts where individuals can strategically modify their inputs to influence outcomes. It categorizes the research into three perspectives: robustness, aiming to design models resilient to "gaming"; fairness, analyzing the societal impacts of such systems; and improvement/causality, recognizing situations where strategic actions lead to genuine personal or societal improvement. The paper introduces a unified framework encapsulating models for these perspectives, including offline, online, and causal settings, and highlights key challenges such as differentiating between gaming and improvement and addressing heterogeneity among agents. By synthesizing findings from diverse works, we outline theoretical advancements and practical solutions for robust, fair, and causally-informed incentive-aware ML systems.