Geometry-Aware Instrumental Variable Regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kremer, Heiner, Schölkopf, Bernhard
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913356888145920
author Kremer, Heiner
Schölkopf, Bernhard
author_facet Kremer, Heiner
Schölkopf, Bernhard
contents Instrumental variable (IV) regression can be approached through its formulation in terms of conditional moment restrictions (CMR). Building on variants of the generalized method of moments, most CMR estimators are implicitly based on approximating the population data distribution via reweightings of the empirical sample. While for large sample sizes, in the independent identically distributed (IID) setting, reweightings can provide sufficient flexibility, they might fail to capture the relevant information in presence of corrupted data or data prone to adversarial attacks. To address these shortcomings, we propose the Sinkhorn Method of Moments, an optimal transport-based IV estimator that takes into account the geometry of the data manifold through data-derivative information. We provide a simple plug-and-play implementation of our method that performs on par with related estimators in standard settings but improves robustness against data corruption and adversarial attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2405_11633
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Geometry-Aware Instrumental Variable Regression
Kremer, Heiner
Schölkopf, Bernhard
Machine Learning
Instrumental variable (IV) regression can be approached through its formulation in terms of conditional moment restrictions (CMR). Building on variants of the generalized method of moments, most CMR estimators are implicitly based on approximating the population data distribution via reweightings of the empirical sample. While for large sample sizes, in the independent identically distributed (IID) setting, reweightings can provide sufficient flexibility, they might fail to capture the relevant information in presence of corrupted data or data prone to adversarial attacks. To address these shortcomings, we propose the Sinkhorn Method of Moments, an optimal transport-based IV estimator that takes into account the geometry of the data manifold through data-derivative information. We provide a simple plug-and-play implementation of our method that performs on par with related estimators in standard settings but improves robustness against data corruption and adversarial attacks.
title Geometry-Aware Instrumental Variable Regression
topic Machine Learning
url https://arxiv.org/abs/2405.11633