External validation of machine learning models—registered models and adaptive sample splitting
The proposed design and splitting approach (implemented in the Python package "AdaptiveSplit") may contribute to addressing issues of replicability, effect size inflation, and generalizability in predictive modeling studies.
Authors
- Ulrike Bingel (ORCID: https://orcid.org/0000-0002-9528-3204)
- Raviteja Kotikalapudi (ORCID: https://orcid.org/0000-0003-4604-3367)
- Robert Englert (ORCID: https://orcid.org/0000-0002-6421-576X)
- Tamás Spisák (ORCID: https://orcid.org/0000-0002-2942-0821)
- Giuseppe Gallitto (ORCID: https://orcid.org/0000-0001-5185-0206)
- Bálint Kincses (ORCID: https://orcid.org/0000-0002-0478-6384)
- Kevin Hoffschlag
- Jialin Li (ORCID: https://orcid.org/0000-0003-2539-1055)
Institutions
- Essen University Hospital (DE)
- Max Planck Institute for Human Cognitive and Brain Sciences (DE)
Publication Details
- Journal
- GigaScience
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1093/gigascience/giaf036
- Citations
- 16
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 39.70
Funders
- Deutsche Forschungsgemeinschaft