mlsanityData to 14 Oct 2026

On Leakage in Machine Learning Pipelines

arXiv 2311.04179 ↗First posted 7 Nov 2023cs.LG cs.AIcited by 7

Not on the map. Why?

Abstract (arXiv)

Machine learning (ML) provides powerful tools for predictive modeling. ML's popularity stems from the promise of sample-level prediction with applications across a variety of fields from physics and marketing to healthcare. However, if not properly implemented and evaluated, ML pipelines may contain leakage typically resulting in overoptimistic performance estimates and failure to generalize to new data. This can have severe negative financial and societal implications. Our aim is to expand understanding associated with causes leading to leakage when designing, implementing, and evaluating ML pipelines. Illustrated by concrete examples, we provide a comprehensive overview and discussion of various types of leakage that may arise in ML pipelines.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 15 of its 101 references that are in the map. First posted on arXiv 7 Nov 2023.

Builds on

15 of 101 references.

References · 15

  1. 2018Datasheets for Datasetscited by 1,386
  2. 2018Model Cards for Model Reportingcited by 1,080
  3. 2020On Hyperparameter Optimization of Machine Learning Algorithms: Theory and Practicecited by 182
  4. 2020Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learningcited by 174
  5. 2020Challenges in Deploying Machine Learning: a Survey of Case Studiescited by 152
  6. 2018The Dataset Nutrition Label: A Framework To Drive Higher Data Quality Standardscited by 97
  7. 2021Bias in Machine Learning Software: Why? How? What to do?cited by 93
  8. 2021Cross-validation: what does it estimate and how well does it do it?cited by 65
  9. 2022Leakage and the Reproducibility Crisis in ML-based Sciencecited by 65
  10. 2019Evaluating time series forecasting models: An empirical study on performance estimation methodscited by 53
  11. 2018Deep Learning Predicts Hip Fracture using Confounding Patient and Healthcare Variablescited by 50
  12. 2014Preventing False Discovery in Interactive Data Analysis is Hardcited by 34
3 more
  1. 2016Assessing and tuning brain decoders: cross-validation, caveats, and guidelinescited by 30
  2. 2021A Too-Good-to-be-True Prior to Reduce Shortcut Reliancecited by 7
  3. 2022Confound-leakage: Confound Removal in Machine Learning Leads to Leakagecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 4.

7 papers.

  1. 2026Radiomics in Medical Imaging: Methods, Applications, and Challengescited by 1
  2. 2025Outlier Detection in Plantar Pressure: Human-Centered Comparison of Statistical Parametric Mapping and Explainable Machine Learningcited by 0
  3. 2025ML2B: Benchmarking LLMs on Cross-Lingual ML Pipeline Generationcited by 0
  4. 2025A geometric and deep learning reproducible pipeline for monitoring floating anthropogenic debris in urban rivers using in situ camerascited by 0
  5. 2026Which Leakage Types Matter? A Quantitative Landscape Across 2,047 Benchmark Datasetscited by 0
  6. 2026Data Leakage in Automotive Perception: Practitioners' Insightscited by 0
  7. 2026MedGate-Fusion: Integrating First-Encounter Semantic Narratives and Physiological Biomarkers for Prospective Stroke Risk Stratificationcited by 0