Contamination Map Methodology

Tabular / Dataset

healthcare_insurance_expenses

regression task, 1338 rows, 7 features (TabArena-v0.1 curated metadata).

Also known as
OpenML 46931, OpenML task 363675
Domain
medical & healthcare
License
Database Contents License (DbCL) v1.0
Homepage
www.openml.org

Exposure by model

BenchmarkModelLevelReason
TabArena LimiX-2 Not seen absent from the complete list of training data
TabArena LimiX Not seen absent from the complete list of training data
TabArena Mitra Not seen absent from the complete list of training data
TabArena Mitra-v2 Not seen absent from the complete list of training data
TabArena Real-TabPFN-2.5 Not seen the authors state the dataset was excluded from training
TabArena Real-TabPFN Unknown training data not disclosed in enough detail
TabArena TabDPT-Turbo (TabDPT v1.2) Not seen the authors state the dataset was excluded from training
TabArena TabDPT v1.3 Unknown training data not disclosed in enough detail
TabArena TabDPT Unknown training data not disclosed in enough detail
TabArena TabICL v1.1 Not seen absent from the complete list of training data
TabArena TabICL Not seen absent from the complete list of training data
TabArena TabICLv2 Not seen absent from the complete list of training data
TabArena TabPFN-2.5 Not seen absent from the complete list of training data
TabArena TabPFN-2.6 Not seen absent from the complete list of training data
TabArena TabPFN-3.5 Not seen absent from the complete list of training data
TabArena TabPFN-3 Not seen absent from the complete list of training data
TabArena TabPFN v2 Not seen absent from the complete list of training data

Notes

TabArena re-upload on OpenML (dataset 46931, task 363675). Original source per TabArena metadata: Kaggle (https://www.kaggle.com/datasets/arunjangir245/healthcare-insurance-expenses/).

Sources