Other

ARCENE Serum Cancer

ARCENE combines mass spectrometry measurements of protein abundance in human serum with synthetic distractor features added for a feature selection challenge. The task is to distinguish cancer samples from normal samples using the publicly labelled data.

Binary classificationLicence: CC BY 4.0OpenML-1458Target: Classopenml ↗

At this operating point

Model comparison

p=10,000, n=100 · Higher Macro-F1 is better

Random Forest is marked with a diamond. Models are ordered by the selected metric.

Across sample budgets

Learning curves

Models

All metric values at this operating point

Model	Macro-F1	MCC	Balanced accuracy	ROC-AUC
Kumo Tabular (M)	0.8456	0.6998	0.8451	0.9245
TabPFN 3.5	0.8357	0.6882	0.8399	0.9372
TabDPT	0.8332	0.6775	0.8393	0.8876
LimiX2	0.8182	0.6515	0.8162	0.9158
RealTabPFN 2.5	0.8066	0.6194	0.809	0.9257
AutoGluon	0.7938	0.6058	0.7979	0.8773
TabPFN Wide 5k (ne3)	0.7918	0.6026	0.7942	0.9108
Causilo	0.7916	0.6048	0.7919	0.8974
RealMLP	0.7782	0.5892	0.7826	0.8884
MLP (tuned)	0.7717	0.5676	0.7802	0.8263
RealTabPFN v2	0.7714	0.576	0.7743	0.8521
MLP	0.7609	0.5564	0.7666	0.8424
Extra Trees (tuned)	0.7596	0.5366	0.7624	0.8571
GeneICL	0.7592	0.5377	0.7655	0.8391
XGBoost (tuned)	0.7574	0.5287	0.758	0.8589
TabM	0.7498	0.5167	0.7509	0.8479
Extra Trees	0.7418	0.502	0.7461	0.8532
TabFM	0.7392	0.5461	0.7413	0.9037
Random Forest (tuned)	0.7387	0.4893	0.7418	0.854
Random Forest	0.7361	0.489	0.7411	0.8614
MITRA	0.7357	0.4925	0.7371	0.8553
Logistic Regression	0.7154	0.4543	0.7185	0.8276
Logistic Regression (tuned)	0.7101	0.4412	0.7123	0.8184
XGBoost	0.7069	0.4266	0.7087	0.8274