Prospect Theory Framework for Behaviorally Realistic Strategic Classification

A new research paper on arXiv (2605.19674) introduces the behaviorally realistic strategic classification problem, addressing the gap between traditional strategic classification models that assume strictly rational agents and real-world decision-making influenced by cognitive biases. The authors propose the Prospect-Guided Strategic Framework (Pro-SF), grounded in prospect theory, to model and learn under behaviorally realistic strategic responses where agents' manipulations deviate from full rationality due to psychological biases. The work draws on evidence from behavioral economics and psychology to formalize this limitation.