Beyond Rational Illusion: Behaviorally Realistic Strategic Classification

2026-05-19

AI Summary

This work proposes a prospect-guided strategic framework (Pro-SF) that integrates prospect theory into strategic classification to better capture human cognitive biases absent in traditional models assuming full rationality. By explicitly modeling three core behavioral mechanisms—gain–loss asymmetry, subjective reference points, and probability distortion—Pro-SF constructs a more realistic model of boundedly rational agents. The framework synergistically combines prospect theory with Stackelberg game dynamics to design a learnable, behaviorally grounded strategic response mechanism. Experimental results demonstrate that Pro-SF significantly enhances robustness and reliability in real-world scenarios, outperforming existing approaches on both synthetic and real-world datasets.

Abstract

Strategic classification(SC) studies the interaction between decision models and agents who strategically manipulate their features for favorable outcomes. Existing SC frameworks typically rely on the idealized assumption that agents are strictly rational. However, evidence from behavioral economics and psychology consistently shows that real-world decision-making is often shaped by cognitive biases, deviating from pure rationality. To formalize this limitation, we identify and define a new problem setting, termed the behaviorally realistic strategic classification problem, where agents' strategic manipulations deviate from full rationality due to psychological biases. Motivated by the identified limitation, we propose the Prospect-Guided Strategic Framework (Pro-SF) to address the problem, a principled framework grounded in prospect theory to model and learn under behaviorally realistic strategic responses. Specifically, to capture behaviorally realistic strategic manipulations, our framework reformulates the Stackelberg-style interaction between agents and the decision-maker by incorporating three key mechanisms inspired by prospect theory, including the asymmetry between benefits and costs, different subjective reference points, and non-rational probability distortion. Experiments on synthetic and real-world datasets establish Pro-SF as a behaviorally grounded approach to strategic classification, bridging machine learning and behavioral economics for more reliable deployment in the real world.

Problem

• strategic classification
• behavioral realism
• cognitive biases
• prospect theory
• non-rational decision-making

Innovation

• strategic classification
• prospect theory
• behavioral realism
• cognitive bias
• Stackelberg game