The ICML 2023 Ranking Experiment: Examining Author Self-Assessment in ML/AI Peer Review

TL;DR

This study analyzes how author-provided rankings during ICML 2023 can be used to calibrate review scores, improving the accuracy of peer review assessments and supporting conference decision processes.

Contribution

It introduces an empirical analysis of author rankings, demonstrating their effectiveness in calibrating review scores and proposes practical applications for conference review management.

Findings

Author rankings improve score calibration accuracy.

Calibrated scores outperform raw review scores in estimating true quality.

Practical applications include aiding senior chairs and award decisions.

Abstract

We conducted an experiment during the review process of the 2023 International Conference on Machine Learning (ICML), asking authors with multiple submissions to rank their papers based on perceived quality. In total, we received 1,342 rankings, each from a different author, covering 2,592 submissions. In this paper, we present an empirical analysis of how author-provided rankings could be leveraged to improve peer review processes at machine learning conferences. We focus on the Isotonic Mechanism, which calibrates raw review scores using the author-provided rankings. Our analysis shows that these ranking-calibrated scores outperform the raw review scores in estimating the ground truth ``expected review scores'' in terms of both squared and absolute error metrics. Furthermore, we propose several cautious, low-risk applications of the Isotonic Mechanism and author-provided rankings in peer review, including supporting senior area chairs in overseeing area chairs' recommendations, assisting in the selection of paper awards, and guiding the recruitment of emergency reviewers.