Microfoundation Inference for Strategic Prediction
This addresses the challenge of social impacts in performative prediction for machine learning practitioners, though it is incremental as it builds on existing concepts with a new estimation approach.
The paper tackles the problem of performative prediction, where predictive models influence the target variable distribution due to stakeholder actions, by proposing a method to learn the distribution map that captures long-term impacts. It models agent responses as cost-adjusted utility maximization, uses optimal transport to align pre- and post-model distributions, and demonstrates results with a convergence rate and empirical validation on a credit-scoring dataset.
Often in prediction tasks, the predictive model itself can influence the distribution of the target variable, a phenomenon termed performative prediction. Generally, this influence stems from strategic actions taken by stakeholders with a vested interest in predictive models. A key challenge that hinders the widespread adaptation of performative prediction in machine learning is that practitioners are generally unaware of the social impacts of their predictions. To address this gap, we propose a methodology for learning the distribution map that encapsulates the long-term impacts of predictive models on the population. Specifically, we model agents' responses as a cost-adjusted utility maximization problem and propose estimates for said cost. Our approach leverages optimal transport to align pre-model exposure (ex ante) and post-model exposure (ex post) distributions. We provide a rate of convergence for this proposed estimate and assess its quality through empirical demonstrations on a credit-scoring dataset.