Paper Reproducibility in ML: A Checklist From 30 Replication Attempts

Missing hyperparameters, unreported data cleaning steps, and hardware-dependent results make ML reproducibility harder than most papers admit.

Overview

This note is part of the field-notes archive generated for this site. The summary below is the published excerpt; you can expand the full write-up anytime in the CMS.

Series

Part of ML in Production (installment 3).

Related notes

• Quantization War Stories Llm
• Ml Drift Monitoring Production

Tags

• machine-learning
• reproducibility
• research
• mlops