Microfoundation Inference for Strategic Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bracale, Daniele, Maity, Subha, Polo, Felipe Maia, Somerstep, Seamus, Banerjee, Moulinath, Sun, Yuekai
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910908693872640
author Bracale, Daniele
Maity, Subha
Polo, Felipe Maia
Somerstep, Seamus
Banerjee, Moulinath
Sun, Yuekai
author_facet Bracale, Daniele
Maity, Subha
Polo, Felipe Maia
Somerstep, Seamus
Banerjee, Moulinath
Sun, Yuekai
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2411_08998
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Microfoundation Inference for Strategic Prediction
Bracale, Daniele
Maity, Subha
Polo, Felipe Maia
Somerstep, Seamus
Banerjee, Moulinath
Sun, Yuekai
Machine Learning
Methodology
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.
title Microfoundation Inference for Strategic Prediction
topic Machine Learning
Methodology
url https://arxiv.org/abs/2411.08998