# Identifying Statistical Bias in Dataset Replication

Source: [Identifying Statistical Bias in Dataset Replication](<https://www.sota2.com/research/paper/68fa684c1e209131cea1b377>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:34:00.401Z

Authors: Logan Engstrom, Andrew Ilyas, Shibani Santurkar, Dimitris Tsipras, Jacob Steinhardt, Aleksander Madry

Published: 2020-05-19T17:48:32.000Z

## Abstract

Dataset replication is a useful tool for assessing whether improvements in test accuracy on a specific benchmark correspond to improvements in models' ability to generalize reliably. In this work, we present unintuitive yet significant ways in which standard approaches to dataset replication introduce statistical bias, skewing the resulting observations. We study ImageNet-v2, a replication of the ImageNet dataset on which models exhibit a significant (11-14%) drop in accuracy, even after controlling for a standard human-in-the-loop measure of data quality. We show that after correcting for the identified statistical bias, only an estimated $3.6\\% \\pm 1.5\\%$ of the original $11.7\\% \\pm 1.0\\%$ accuracy drop remains unaccounted for. We conclude with concrete recommendations for recognizing and avoiding bias in dataset replication. Code for our study is publicly available at http://github.com/MadryLab/dataset-replication-analysis .

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2005.09619>)

[PDF](<https://arxiv.org/pdf/2005.09619.pdf>)