Estimating Misreporting in the Presence of Genuine Modification: A Causal Perspective
This addresses a fundamental challenge in strategic ML settings for policymakers and system designers, though it is incremental as it builds on prior work on strategic responses.
The paper tackles the problem of distinguishing strategic misreporting from genuine feature modification in ML-based resource allocation, proposing a causal approach that identifies and quantifies misreporting rates by exploiting asymmetry in causal effects on downstream variables, with empirical validation on semi-synthetic and real Medicare datasets.
In settings where ML models are used to inform the allocation of resources, agents affected by the allocation decisions might have an incentive to strategically change their features to secure better outcomes. While prior work has studied strategic responses broadly, disentangling misreporting from genuine modification remains a fundamental challenge. In this paper, we propose a causally-motivated approach to identify and quantify how much an agent misreports on average by distinguishing deceptive changes in their features from genuine modification. Our key insight is that, unlike genuine modification, misreported features do not causally affect downstream variables (i.e., causal descendants). We exploit this asymmetry by comparing the causal effect of misreported features on their causal descendants as derived from manipulated datasets against those from unmanipulated datasets. We formally prove identifiability of the misreporting rate and characterize the variance of our estimator. We empirically validate our theoretical results using a semi-synthetic and real Medicare dataset with misreported data, demonstrating that our approach can be employed to identify misreporting in real-world scenarios.